CURRENCY AND EQUIVALENT UNITS Currency Unit US$1= 18,0774 (As of June 15, 2015) FISCAL YEAR January 1 – December 31 WEIGHTS AND MEASURES The metric system is used through report ACRONYMS AND ABBREVIATIONS AIPA Agency of Interventions and Payment for IAMO Insitut fur Agrarentwicklung in Mittel- and Agriculture Osteuropa AMA Association Membership Agreement ISPA Instrument for Structural Policies for Pre Accession ANSA National Food Safety Agency MAFI Ministry of Agriculture and Food Industry BOOST Government Budget and Expenditure Data MTEF Medium-Term Expenditure Framework CEE Central and Eastern Europe NMS New Member States CAP Common Agricultural Policy NARDS National Agriculture and Rural Development Strategy CIT Corporate Income Tax OECD Organization for Economic Co-operation and Development CIS Commonwealth of Independent States PER Public Expenditure Review CNAS National Office of Social Insurance PHARE The Programme of Community aid to the countries of Central and Eastern Europe DCFTA Deep and Comprehensive Free Trade Area PIT Personal Income Tax ECA Europe and Central Asia PP Percentage Point EU European Union R&D Research and Development GDP Gross Domestic Product SAPARD Special accession programme for agriculture and rural development GoM Government of Moldova SSCs Social Security Contributions HVA High-Value Agriculture STOA Science and Technology Options Assessment IFIS International Financing Institutions TFP Total Factor Productivity Regional Vice President: Laura Tuck Country Director: Qimiao Fan Senior Director: Marcelo Giugale Practice Manager: Ivailo Izvorski Task Team Leader: Emilia Skrok, Luc Razafimandimby ii Table of Contents ACRONYMS AND ABBREVIATIONS ................................................................................................................ ii Acknowledgements....................................................................................................................................... 1 Executive summary ....................................................................................................................................... 2 Chapter 1. Challenges to Agricultural Development ............................................................................ 10 1.1. Economic Perspective ................................................................................................................. 10 1.2. Policy Perspective ....................................................................................................................... 16 1.3. Moldova’s Fiscal Challenges and Public Support to Agriculture ................................................. 18 1.4. Conclusions ................................................................................................................................. 21 Chapter 2. Level and Composition of Government Spending on Agriculture ...................................... 23 2.1. Trends in the Budget for Agriculture ............................................................................................... 23 2.2. Key Aspects of Economic and Functional Composition ................................................................... 25 2.3. Effectiveness of Public Spending in Supporting the National Sector Strategy ................................ 28 2.4. Distributional Impact and Inequality of Agricultural Subsidies ....................................................... 36 2.5. Conclusions and Recommendations ................................................................................................ 37 Chapter 3. Tax Expenditures in Agriculture .......................................................................................... 39 3.1. Taxation of Agriculture in Moldova ................................................................................................. 39 3.2. The Fiscal Cost of Agricultural Tax Expenditures ............................................................................. 42 3.3. The Economic and Social Impact of Tax Expenditure in Agriculture ............................................... 44 3.4. What Might be Changed? ................................................................................................................ 48 Conclusions and Recommendations ....................................................................................................... 52 Annexes ................................................................................................................................................... 54 References .................................................................................................................................................. 78 TABLES Table 2.1: Structure of General Government Expenditure by Function, 2009–13 (Percent) ..................... 24 Table 2.2: Recurrent vs Capital Agriculture Spending ................................................................................ 25 Table 2.3: Economic Structure of Recurrent Spending (Percent of GDP) ................................................... 25 Table 2.4: Functional Classification of Executed Budgets, 2009–13 (Percent of total spending) .............. 26 Table 2.5: Functional Classification of Executed Budgets for Services and Subsidies, 2009–13 ................ 28 Table 2.6: Sector Priorities and Allocation of Agriculture Spending ........................................................... 29 iii Table 2.7: Risk Management Package (RMP) 2014 (Millions of US dollars) ............................................... 34 Table 2.8: Main Risks Insured in Moldova, 2011–2014 (Percent) .............................................................. 35 Table 3.1: Value-Added Tax (VAT) Rates..................................................................................................... 42 Table 3.2: Estimated Cost of Tax Expenditures in Agriculture .................................................................... 43 Table 3.3: What Is Taxed Where, Moldova and Regional Comparators ..................................................... 46 Table 3.4: Budgetary Effect, Proposed Changes to Tax Concessions in Agriculture ................................... 52 FIGURES Figure 1-1: Growth of Moldova’s GDP and Gross Value Added (GVA) in Agriculture, 2001–14 (Percent change over previous year)......................................................................................................................... 11 Figure 1-2: Moldova’s GDP Growth Decomposition, 2009–14 (Percentage points) .................................. 11 Figure 1-3: Share of Agriculture GVA in Total GVA, 2002–12 (Percent) ..................................................... 12 Figure 1-4: Agriculture GVA vs. Income Level (in percent and constant US dollars 2005) ......................... 12 Figure 1-5: Employment in Agriculture, 2002–12 (Percent of total employment) ..................................... 13 Figure 1-6: Agriculture GVA vs. Income Level, 2000–12 (in percent and PPS) ........................................... 13 Figure 1-7: GVA per Worker by Sectors Relative to Economy Wide Average (Percent)............................. 14 Figure 1-8: Public Spending and Capital Stock per Hectare ........................................................................ 14 Figure 1-9: Effect of Weather on Prices of Crops and Livestock, 1992–2012 (Current prices, MDL, 2000=100) ................................................................................................................................................... 15 Figure 1-10: Agriculture GVA Growth Decomposition, 2002–13 (Percentage points) ............................... 15 Figure 1-11: Value of Moldova's Agro-food Exports (Million USD) ............................................................ 16 Figure 1-12: Destinations for Moldova's Agro-food Exports (Share in total, percent) ............................... 16 Figure 1-13: Government Spending Structure, Moldova, (Percent of GDP)............................................... 19 Figure 1-14: Government Spending vs. GDP per Capita, 2012 (in percent of GDP and US dollars, PPP) .......................................................................................................... 19 Figure 1-15: Tax Revenue Structure, Moldova, (Percent of GDP) .............................................................. 19 Figure 1-16: Tax Revenue Structure, 2012 (Percent of GDP)...................................................................... 19 Figure 1-17: Government Spending on Agriculture (Percent of GDP) ........................................................ 20 Figure 1-18: Agriculture: Government Spending on Agriculture and Total Factor Productivity (Free Disposal Hull Method) ................................................................................................................................ 21 Figure 1-19: Agriculture: Government Spending on Agriculture and Total Factor Productivity (Batesse and Coelli (1988) Method) .......................................................................................................................... 21 Figure 2-1: Moldova: Public Spending and Spending on Agriculture, 2002–14 (Percent of GDP) ............. 24 Figure 2-2: Agriculture Budget, 2009–13, (Percent of GDP) ....................................................................... 24 Figure 2-3: Structure of Domestically Funded Investment in Agriculture, 2009–13, (Percent) ................. 31 Figure 2-4: Distribution of Investment Subsidies, 2012–14, (Million USD) ................................................ 31 Figure 2-5 : Distribution of Recurrent Subsidies, 2012–14 (Millions of US dollars) ................................... 32 Figure 2-6: Corporate and Individual Subsidy Beneficiaries, 2012 (Percent) ............................................. 36 Figure 3-1: Taxes Collected from Agriculture by Tax Type, 2008–14 (Percent of GDP) ............................. 40 Figure 3-2: Investment in Long-Term Tangible Agricultural Assets (Percent of GVA) ................................ 47 Figure 3-3. Benefit of the VAT Discount, by Quintile, 2013 (Percent) ........................................................ 48 iv ANNEXES Annex 1: Modernization and restructuring of agri-food industry and adapting it to EU standards – the experience of Poland. ................................................................................................................................. 54 Annex 2: Determinants of TFP in agriculture .............................................................................................. 55 Annex 3: Local mapping of Agriculture Expenditures ................................................................................. 58 Annex 4: Agricultural Support 2014 ............................................................................................................ 59 Annex 5: Agricultural Education and Research Institutions........................................................................ 61 Annex 6: Taxation of Agriculture Property in Moldova .............................................................................. 64 Annex 7: Preferential Tax Treatment in Agriculture ................................................................................... 65 Annex 8: Calculating VAT GAP .................................................................................................................... 67 Annex 9: Concessions in Taxes on Incomes (unincorporated businesses) ................................................. 68 Annex 10: Fundamental Principles of a Fair and Effective Tax System ...................................................... 70 Annex 11: Reduced VAT rates for food and agricultural products – pros and cons ................................... 71 Annex 12: Distributional Impact of a VAT Increase for Agricultural Products ............................................ 74 Annex 13: Tax Incentives and Investment in SMEs..................................................................................... 76 v Acknowledgements This Public Expenditure Review reviews Moldova’s public expenditures and tax policy in the agricultural sector with an emphasis on the period 2009-2013. This study was prepared by a team comprising Ruslan Piontkivsky (Senior Economist, GMFDR), Emilia Skrok (Senior Economist, GMFDR), Luc Razafimandimby (Senior Economist, GMFDR) and Felicia Pricop (Operations Officer, GTCDR), with support from Irina Capita (Consultant, GMFDR), Julie Sati Lohy (Consultant, GMFDR), Marcel Chistruga (Consultant, GMFDR), Jan Gąska (Consultant, GMFDR), Ewa Korczyc (Country Economist, GMFDR), Mikhail Matystin (Consultant, GPVDR), Caterina Ruggeri Laderchi (GPVDR) and Irina Schuman (Senior Agricultural Economist, GFADR). Special thanks to Agnieszka Boratyńska (Consultant, ECCPL), Mismake Galatis (Program Assistant, GMFDR) and Viorica Strah (Program Assistant, ECCMD) for their outstanding assistance. The team would like to thank the authorities for the excellent collaboration throughout the preparation of the Public Expenditure Review and the valuable advice and inputs provided to complete the report. The assistance, cooperation, and input of many government departments in Moldova is gratefully acknowledged. They are too numerous to mention individually. The team is greatly indebted to peer reviewers: Peter Siegenthaler (Senior Economist, GMFDR), Theo David Thomas (Lead Economist, GMFDR) for their invaluable comments. The report was completed under the leadership of Ivailo Izvorski (Practice Manager, GMFDR). The team benefited from the valuable advice of Anatol Gobjila (Senior Operations Officer, GAFDR), Satu Kristiina Kahkonen (Director, GMFDR), Alexander Kremer (Country Manager, ECCMD) and Lalita Moorty (Program Leader, ECCU2). 1 Executive summary Agriculture is critical to Moldova’s economy, but the sector faces significant challenges. Agriculture is vital to Moldova’s economy. It accounts for 14 percent of GDP and employs 25 percent of the labor force; together with agro-processing, it generates more than 20 percent of GDP. While the size of the agriculture sector is one of the largest in Europe, it is comparable to other countries at similar income levels. The sector is also a major foreign exchange earner, with agro-food products comprising about 60 percent of merchandise exports. While the share of agriculture in value-added and employment has decreased over the years in Moldova, the sector continues to be both a driver of growth and an avenue for reducing poverty. The Government of Moldova (GoM) identifies agriculture as one of the main drivers of growth in its 2020 National Development Strategy. The GoM’s strategic vision is to modernize agriculture to improve competitiveness and better integrate producers into markets. Improving competitiveness becomes even more important as Moldova seeks to increase agricultural exports to the European Union (EU) under the Association Agreement and the Deep and Comprehensive Free Trade Area (DCFTA)—a reorientation that is becoming all the more pressing because of deepening uncertainties about Moldova’s traditional markets in the Commonwealth of Independent States (CIS). A review of the experience of New EU Member States shows that agriculture benefited after accession—with the benefit depending not on the level of public support but on how it was provided. Countries that focused public resources on excessive price and market support did not gain as much as those that modernized the agro-food sector and adopt EU standards. Agriculture in Moldova faces both short-and-long term challenges. While the share of agriculture in value-added and employment is significant, productivity is low at about 60 percent of average national productivity. Despite recent improvements, agriculture’s share in employment is twice that in GDP – implying that labor generates less valued-added in agriculture than in other sectors. Given this low productivity, agricultural incomes remain modest, leading to much higher poverty incidence in rural areas (18.8 percent) than in urban areas. Moreover, as a small open economy, real GDP growth in Moldova is volatile because external economic shocks are more easily transmitted to the domestic economy. This high exposure to external shocks is amplified by climate-related shocks that affect agriculture; recurring droughts, coupled with weak risk-mitigation mechanisms, periodically suppress agricultural output, thus adding to the volatility of overall growth. These climate shocks are projected to become more severe in the next 40 years. Finally, recent bans on agricultural exports to Russia are negatively affecting Moldova’s agricultural exports and exacerbating growth volatility. Increasing public spending on agriculture is not an optimal response to these challenges. Instead, to make the sector more competitive and resilient to shocks, direct budget spending should be reduced in some areas and redesigned or increased in others. Tax expenditures could be streamlined to make them more efficient, cost effective, and equitable. 2 In Moldova, public spending on agriculture is through direct budget spending and tax expenditures (foregone tax revenues). Direct budget spending is in the form of subsidies (current and capital) and services (research, education, food safety, and extension services) and through donor-supported programs. Tax expenditures are through reduced corporate income tax (CIT), value-added tax (VAT), social security contributions (SSCs), and health contribution rates. Tax concessions, a type of tax expenditure, take the form of reduced tax rates, exemptions, and tax deductions. Direct budget spending at 1.4 percent of GDP is high in Moldova compared to other countries, and public support reaches about 2 percent of GDP when tax expenditures are added. Moldova spent 1.4 percent of GDP and 3.6 percent of total government outlays on agriculture in 2013— higher than many countries in Europe and Central Asia (ECA), including new EU member states like Romania and Bulgaria.1 Moldova also spent 0.6 percent of GDP in the form of foregone taxes. Further increases in public spending are not warranted, especially given overall fiscal considerations. As in many small open economies, Moldova has a relatively large public sector. Direct budget spending on agriculture tripled during 2002–14.2 Even though public resources devoted to agriculture have increased, the sector’s productivity has not gone up much. To ensure fiscal sustainability, government needs to continue fiscal consolidation to reduce the deficit to 3 percent GDP by 2016. In the medium term, the government would need to balance demographic spending pressures and pressing needs for investment to upgrade obsolete infrastructure and reinforce the human capital base. It is thus neither possible nor necessary to increase public spending on agriculture any more than current levels. So the challenge for Moldova is to ensure that its direct budget spending is well-used – reduced in some cases, and redesigned or increased in others. Meanwhile, tax expenditures can be made more efficient, cost effective, and equitable. With respect to direct budget spending, allocative efficiency has improved, but concerns about efficiency and equity remain. Allocative efficiency in direct budget spending on agriculture has improved. Spending is aligned with the government priorities reflected in the three main pillars of the National Agriculture and Rural Development Strategy: (1) enhanced competitiveness of the agricultural and agro-food sector through restructuring and modernization; (2) sustainable management of natural resources; and (3) better conditions for living and working in agricultural and rural areas (Table ES.1). More than 80 percent of direct budget spending on agriculture is directed to the first priority. In particular, a large and increasing share of capital spending in the budget is directed to upgrading obsolete infrastructure. About 30 percent of the direct budget spending in this pillar is for core services to support agricultural competiveness, such as research, education, and extension. Under Pillar II, support for sustainable management of natural resources accounts for less than 15 percent of expenditures; here the largest allocation is for hail prevention systems. As yet the aspiration of better rural living and working 1 3.6 percent is using COFOG for international comparison because it includes fishery and forestry. When it is limited to local mapping, taking into account a narrower definition, it is 2.6 percent. In both cases, it is high. 2 Moldova’s comparator countries reduced spending during that period. 3 conditions is not well- reflected in the agricultural budget; what little support this area gets is mainly provided by donors. In any case, Pillar III involves a wider set of ministries, including agriculture. Table: The National Agricultural and Rural Development Strategy Percent Amount of Total Priority Spending Category (MLD) Spending Pillar I. Enhanced competitiveness through restructuring and modernization 1,140,712,526 82.8 Investment subsidies: machinery and equipment, livestock equipment and breeds, post-harvest and 343,982,861 25.0 Component A: Increased investments for processing equipment (M3, M5, M6, M7, M8) the modernization Donor financed projects (WB, IFAD, MCC, EIB) 369,237,736 26.8 Component B: Improved access to capital Current subsidies: credit and land consolidation 39,341,478 2.9 and input and output markets for farmers (M1, M9) Component C: Enhanced agricultural education, research, and extension Agri Services : food safety, agriculture education, services and strengthened synergies research, extension, testing and registration of new 388,150,451 28.2 among them; reformed food safety crop varieties, animal identification system Pillar 2. Sustainable management of natural resources 171,291,890 12.4 Current subsidies: risk insurance, energy for irrigation (M2, M10); Capital subsidies: irrigation equipment, anti-hail and anti-frost systems, 70,631,072 5.1 Component A: Reduced climate-related protected field vegetable production (part of M3, risks for agricultural production M5, the whole M4) Agri Services: irrigation and anti-hail 100,660,818 7.3 Pillar 3. Improved conditions for living and working in agricultural and rural areas 65,159,600 4.7 Donor-financed projects 65,159,600 4.7 Pillar I, enhancing agricultural competitiveness through restructuring and modernization, is supported by sizable and well-targeted capital spending, mainly in the form of donor programs and investment support/subsidies to producers. In 2013 capital spending constituted about 60 percent of total spending on agriculture in Moldova—almost twice as much as in 2010 and far above the 10–30 percent in comparator countries. Most of the capital spending supports Pillar 1 of the government strategy. The increase in capital spending in Moldova reflects much better execution of the budget, inflows of donor funds for investment, and a shift from recurrent to investment subsidies. Investment subsidies 4 constitute about 25 percent of spending in support of the competitiveness agenda of agriculture in Moldova and are used efficiently to address a major need: modern and innovative production and post- production practices in agriculture, especially in the high-potential high-value-added sector. The investment subsidies support programs that respond to the modernization needs of the sector. They have also moved away from a concentration on machinery to smart subsidies that promote technology. In contrast, recurrent subsidies intended to strengthen agricultural competitiveness (such as credit subsidies or the land consolidation program) may need to be redesigned. Uptake of land consolidation programs is very limited and credit is allocated primarily to wealthier farmers and subsidizes agricultural inputs. The cost-efficiency and effectiveness of most services supporting agricultural competiveness are debatable. Government support has been sizable and stable, with about 30 percent of agricultural spending allocated to research, education, food safety, and extension services. International studies have proved that these services deliver solid returns on public investments and boost agricultural productivity and competiveness. However, in Moldova spending on agriculture research and education is driven by a sizable wage bill and the institutional structure is obsolete, with limited ability to generate outcomes (quality of education or research spillovers to extension services). For instance, research and the curricula at agricultural colleges do not respond to farmers’ needs. There are also issues related to the limited funding for extension services, which have proved efficient and popular (accessed by 50 percent of farmers), and the role and results of the Seeds Commission and the National Food Safety Agency (ANSA). Neither amounts nor allocations of funds to programs for sustainable management of resources under Pillar II look suitable to the challenges confronting the sector. Risk management programs account for about 15 percent of total agriculture spending, yet the exceptionally high output volatility, driven by weather-dependent variation in crops and irrigation, are a drag on sector development. Risk is mainly managed by anti-hail services, subsidies for horticulture production and irrigation, and subsidies for risk insurance. The largest allocation goes to hail prevention systems, which have not yet proved to be cost- efficient (World Bank 2007). The results from spending on the irrigation system and insurance subsidy program are not particularly encouraging. The irrigation system is underutilized and funds for maintenance are limited. For instance, the system has the capacity to irrigate over 100,000 hectares of land but annual demand for irrigation services is only about 7,000–15,000 ha). The risk insurance subsidy, which is based on the traditional multiperil insurance system, has been in place since 2006 but has not resulted in extensive coverage and is generally not considered well-suited for managing national systemic risks like drought and frost. There is also anecdotal evidence of fraud related to utilization of insurance subsidy payments.3 Inequities in access relate to the size and legal status of farms; there are also geographic and gender dimensions. A comprehensive analysis of distributional implications of public spending on agriculture is complicated by the lack of data on both small and commercial farms and on the welfare of farm owners. 3 The subsidy is claimed by the insurance company (50 percent of the premium), not the insured farmer. 5 There is some evidence, however, that public support is skewed toward large corporate farms in richer regions and toward male farmers. For the past three years, 75–80 percent of subsidies have been absorbed by large corporate farms. The only subsidy that largely benefits smaller individual farms stimulates investment in protected-field vegetable production. These outcomes are inconsistent with the weight of individual farmers—70 percent—in total agricultural output. Small farmers also seem to be more in need of public support because they face greater information and monetary constraints in accessing modern technologies. Determined improvement of the efficiency and effectiveness of Moldova’s spending on agriculture would build up both its competiveness and its resilience to shocks. Three types of government action on current programs could make spending more efficient and effective: reduce, redesign, or increase funding.  The authorities should consider reducing spending for the credit subsidy program and the land consolidation program (if uptake remains small), and the anti-hail service.  Programs that call for redesign or further reform are research; the agriculture education system (institutional reform); management of the irrigation system (a new model for functioning of the water agency and the water user associations); and support for modernization of the food safety system.  Programs that appear to deserve more funding given their importance to the sector or positive results achieved so far are modern risk management programs, after improvement of current programs; extension services that are generating positive results; subsidized investment in post-harvest infrastructure, funded by a reduction of the large machinery program; and ‘smart’ subsidies aimed at promoting use of technology. Increases in these areas, however, need to be implemented in an unchanged overall envelope given concerns about the government fiscal space. Tax expenditures result in additional generous support but need to be redesigned to more effectively address the challenges in the agriculture sector. Agriculture receives highly favorable tax treatment to compensate for its high risks, low profitability, and high capital intensity. Farmers benefit from such favorable tax arrangements as an alternative to direct spending. The main tax expenditures are reduced rates for CITs, SSCs and health contributions, concessions for citizens who acquire farming property, property tax concessions, and VAT concessions. The fiscal cost associated with this favorable tax treatment is sizable. The annual cost of farming sector benefits from tax arrangements is estimated at more than 0.6 percent of GDP, with the largest share being for VAT treatment, followed by direct taxes and social security contributions. Tax expenditures amount to about 60 percent of tax collections in agriculture and about 40 percent of direct budget spending on agriculture. 6 Moreover, tax expenditures do not effectively address the challenges identified for the sector. The favorable tax treatment of income and real property is likely to slow restructuring of agriculture. Reduced rates for health contributions for individual farmers and reduced tax rates for corporate farmers are likely to affect the decisions of farmers and landowners to stay in the industry and are likely to slow structural transformation of the economy. Moreover, Moldova’s current system for taxing real property, including transfers of agricultural property, may delay exit from agriculture and raise impediments to entry, particularly for foreigners. Also, VAT concessions on agricultural inputs (e.g., investment and tractors) are not justified by market failures and are likely to have an economic cost by inducing changes in the pattern of production. There is also evidence that a reduced VAT rate on selected agricultural products is in general regressive—it accrues mostly to higher-income quintiles. Thus, it is questionable whether the reduced VAT rate is the best instrument for improving income distribution. Tax expenditures could be redesigned to tackle agricultural problems in ways that are more efficient, cost-effective, and equitable. One option would be to pursue a range of both tax and non-tax incentives—or removal of disincentives—related to capital investment, farm consolidation, land use, and income support. The main changes proposed are:  Introduce a capital allowance or a tax credit similar to that used in OECD countries. Finance that by returning the CIT rate for agricultural enterprises to the standard 12 percent and removing the VAT exemption for machinery and tractors.  Provide incentives for better use of land by abolishing the limitation on land acquisition by foreigners; more accurate valuation of land and property, which is vital to adopting value-based taxation; and introducing a special tax on uncultivated land.  Introduce a presumptive turnover-based tax, with a low tax burden and compliance costs, to bring low-income farmers back into the tax net and make means-tested social assistance more effective.  Limit disincentives to move out of agriculture and consider removing the reduction in the SSC rate for farmers.  Reintroduce a unified VAT rate for agricultural products (while ensuring proper compensation for low-income households) to help limit production inefficiencies and administrative costs. Table: Summary of Policy Options I. Spending Policies to increase value for money of public supportfor money of public support 1. Consider reducing spending on programs that have not brought expected results: 7 Cancel the credit subsidy program Redesign or revoke the land consolidation program Replace the anti-hail program with a more efficient risk prevention program 2. Take an action to increase value for money of several spending programs: Improve the quality of agriculture research and education Revisit management of the irrigation system Reform the Seeds Commission testing program Continue to reform the food safety programs 3. Increase budget allocation for the following programs: Build up the extension services Invest more in the post-harvest infrastructure program funded by a reduction of the large machinery program Continue increasing support to ‘smart’ subsidies aimed at promoting use of technology 4. Continue improving the annual budget process Consider introduction of performance evaluations of spending programs II. Tax Policies to tackle agricultural problems in more efficient way 1. Revamp incentives to invest in agriculture: Reconsider current tax incentives for private pension savings Increase the CIT rate for agricultural enterprises to the standard 12 percent Remove the VAT exemption for machinery and tractors 2. Incentivize better use of land: Abolish or ease the restriction on land acquisition by foreigners Continue effective valuation of land and property and adopt value-based taxation of property Introduce a special tax on uncultivated land 3. Provide income support to the poorest farmers: Introduce a presumptive turnover-based tax Consider removing the reduction in the SSC rate for farmers 8 4. Re-introduce a unified VAT rate for agriculture products combined with compensation for low-income households. 9 Chapter 1. Challenges to Agricultural Development The share of Moldova’s agricultural value-added and employment is significant and in line with other countries at comparable income levels. The volatility of agriculture output, mainly caused by extreme weather, is a major risk factor for agriculture growth and poverty reduction in rural areas. While the impact of extreme weather cannot be fully mitigated, better land utilization, more efficient irrigation, and more productive and efficient government outlays on agriculture can help to stabilize agricultural production. Moldova’s public support to agriculture is higher than in many other countries at a similar level of development. Moldova’s public support is about 2 percent of GDP, including 0.6 percent of GDP in tax expenditures. Clearly, fiscal policy is a key factor in agricultural policy but the government does not have the space for further increases in spending. Meanwhile, budget preparation and monitoring is yet to become an effective way to manage public interventions in agriculture. The rest of this chapter is organized as follows. The first part looks at Moldova’s agriculture sector from an international perspective and provides background on recent trends and developments. The second part focuses on sectoral policies in agriculture and government priorities for the sector. The third part outlines the fiscal envelope and previews the size of public support to the agriculture. The fourth part summarizes conclusions of the chapter. 1.1. Economic Perspective 1. Moldova’s recent economic performance has been strong but dependent on the evolution of the agricultural sector. Moldova has registered the highest cumulative growth among regional peers, almost 32 percent, from the pre-crisis level of 2007 through 2014. However, as a small open economy with a large agricultural sector, its growth has been volatile because of its vulnerability to climatic and global economic conditions. As Figure 1-1 and Figure 2 indicate, to a large extent recent GDP volatility has been driven by agriculture. 2. Moldova’s agriculture sector is one of the largest in Europe. It generates about 12 percent of gross value-added.4 Neighbors in the Commonwealth of Independent States (CIS), countries in the Western Balkans, and some new EU new member states are good comparators for Moldova because they have a common external environment. In the global context however, countries like Guatemala, the Philippines, Honduras, and Bolivia have very similar economic structures (Box 1.1). The large share of value-added reflects the high quality of Moldova’s agriculture land, which is considered the country’s main natural resource. In terms of nature and geography, Moldova has considerably greater potential for agriculture than many larger European countries. Total 4 When agriculture and agro-processing are combined, the importance of agriculture to the Moldovan economy as a combined sector accounts for about 17 percent of GDP 10 agricultural land, about 2.5 million ha, represents almost 73 percent of its territory.5 Arable land accounts for 1.84 million ha and perennial plantations for 0.3 million ha. Figure 1-1: Growth of Moldova’s GDP and Gross Figure 1-2: Moldova’s GDP Growth Value Added (GVA) in Agriculture, 2001–14 Decomposition, 2009–14 (Percent change over previous year) (Percentage points) 60 10 Other services 50 8 40 Finanical intermediation 6 services 30 Net taxes 20 4 10 Transport and 2 communications 0 Construction -10 0 -20 2009 2010 2011 2012 2013 2014 Trade -30 -2 Industry GVA in Agriculture, y/y GDP growth, y/y -4 Source: National statistics. Source: National statistics and World Bank staff 3. Moldova’s agriculture sector is one of the largest in Europe. It generates about 12 percent of gross value-added.6 Neighbors in the Commonwealth of Independent States (CIS), countries in the Western Balkans, and some new EU new member states are good comparators for Moldova because they have a common external environment. In the global context however, countries like Guatemala, the Philippines, Honduras, and Bolivia have very similar economic structures (Box 1-1). The large share of value-added reflects the high quality of Moldova’s agriculture land, considered the country’s main natural resource. In terms of nature and geography, Moldova has considerably greater potential for agriculture than many larger European countries. Total agricultural land, about 2.5 million ha, accounts for 73 percent of its territory.7 Arable land accounts for 1.84 million ha and perennial plantations for 0.3 million ha. 5 Includes lands in the forestry fund. 6 When agriculture and agro-processing are combined, the importance of agriculture to the Moldovan economy as a combined sector accounts for about 17 percent of GDP 7 Includes lands in the forestry fund. 11 Box 1-1: Benchmarking Moldova’s Agriculture Sector Benchmarking Moldova’s economy by taking into account the prominent role of agriculture for both growth and employment makes it possible to assess the country’s performance and identify factors that differentiate it from its peers. Such an exercise, although performed at a highly aggregated level, may be useful for international comparisons, which are a good starting point for sharing knowledge and experience. Moldova’s benchmarks are determined by taking into account the sectoral composition of value-added, employment by sector, and GDP per capita (PPP terms) using a global and European sample. Benchmarks countries in the global sample are Guatemala, the Philippines, Nicaragua, Paraguay, El Salvador, Honduras, Sri Lanka, Georgia, Mongolia, Serbia, Turkey, Bolivia, and Uzbekistan. European benchmarks used in this report are Armenia, Belarus, Georgia, Kazakhstan, Kyrgyz Republic, Tajikistan, Turkmenistan, and Ukraine; Bosnia, Macedonia, and Serbia in the Western Balkans; and EU members Hungary, Croatia, Lithuania, and Romania. Source: WDI data. 4. As in other countries, the structural transformation of Moldova’s economy has led to a decline in the relative importance of agriculture. Over the last 15 years, the share of agriculture in value- added dropped from 29 percent in 2000 to as low as 10 percent in 2009, before bouncing back to 11–13 percent in 2011–14, mostly because of high agricultural prices. At the same time, the share of industry declined by some 2 percentage points (pp) to 14 percent of GDP in 2014. Moldova’s shift away from agriculture and into services is a typical development pattern, but although its relative share of GDP is declining, agriculture continues to exhibit healthy growth rates. Figure 1-3: Share of Agriculture GVA in Total GVA, Figure 1-4: Agriculture GVA vs. Income Level 2002–12 (in percent and constant US dollars 2005) (Percent) Source: WDI data. Source: WDI data. 12 5. Over 25 percent of Moldova’s labor force is employed in agriculture. While significant, this share is lower than among its peers with similar incomes per capita and well below the almost 50 percent share in the late 1990s. The decline in agricultural employment reflects massive emigration of Moldova’s rural workforce that started in the 1990s and is continuing, though at a slower pace. This labor re-allocation episode stands out when compared to both European economies and the broader group of emerging market comparators (Figure 1-5). 6. While productivity in agriculture has been growing recently, it is still moderate, at about 60 percent of average productivity in Moldova—similar to Ukraine and Russia but significantly higher than in Romania or Croatia (Figure 1-7). In recent years, the steep reduction in agriculture employment was one of the main reasons behind the growth in the productivity of Moldovan agriculture. Still, agriculture’s share of employment is more than twice its share of GDP, which implies that labor generates much less value- added in agriculture than it does in other sectors. Figure 1-5: Employment in Agriculture, 2002–12 Figure 1-6: Agriculture GVA vs. Income Level, 2000–12 (Percent of total employment) (in percent and PPS) 70 60 MDA Share of employment in agriculture UKR 60 50 MCD 50 40 ROM GEO 40 30 RUS 30 PHL 20 ECU 20 BIH 10 10 MNG 0 PRY 0 0 5000 10000 15000 20000 25000 SRB MDA UKR ARM AZE BIH ALB BGR KAZ ROM RUS LTU HRV HUN GDP per capita PPP TUR 2002 2012 Source: WDI data Source: WDI data. 7. Low incomes from agriculture results in higher poverty in rural areas. Household income from agriculture has been dropped over the years, from 28.4 percent of total income in 2007 down to 19.1 percent in 2010. Poverty is highest in rural areas, particularly among those who are agriculture-dependent. While in 2013 in urban areas the poverty rate was fairly low, at 4.6 percent, rural poverty was much higher at 18.8 percent. However, agriculture in rural areas is still a safety net protecting the rural poor from food insecurity because agricultural activities facilitate their access to food, while urban households rely primarily on purchased food. 13 Figure 1-7: GVA per Worker by Sectors Relative to Figure 1-8: Public Spending and Capital Stock per Economy Wide Average Hectare (in percent of GDP and US dollars) (Percent) 3.00% 5 4.5 2.50% 4 2.00% 3.5 3 1.50% 2.5 2 1.00% 1.5 0.50% 1 0.5 0.00% 0 Agriculture spending to GDP Capital stock per hectare (thous. 2005 USD) Source: WDI data. Source: WDI data. 8. Investment in agriculture is essential for improving the potential of the sector and helping to mitigate its volatility. In recent years, the increasing stock of capital has had an important role in generating agricultural production. International comparisons suggest that Moldova’s agricultural capital stock is equal to about the average for the group despite relatively high government outlays (Figure 1-8). 9. Land utilization, variation of crop production, irrigation, and government outlays have a significant impact on agricultural production and on total factor productivity (TFP). Large variations in the production of crops, which constitutes over 50 percent of total agriculture output in Moldova, were the key factor behind output volatility in recent years and resulted in a large TFP shift over the period. The analysis of TFP (see Annex 2) also points to a weak, though positive relation of the size of farms to productivity. Increased capital investment in agriculture and a declining rural labor force have also affected agricultural production, but to a much lesser extent than weather conditions. 10. Exceptionally high output volatility poses a serious threat to sector development (Figure 1-9 and 1.10). Crop production is particularly vulnerable to the weather. Severe droughts in 2003, 2007, and 2012 had a devastating effect on most crops.8 High volatility partly reflects underdeveloped weather-related risk mitigation, varying access to irrigation, low adoption of modern agricultural technologies, such as drought-resistant crop varieties and anti-hail protection tools, and a lack of 8 The severe drought in 2007 caused a 55 percent decline in the winter wheat harvest, a 72 percent decline in maize, a 70 percent decline in sunflower seeds, a 64 percent decline in pea production, and a one third reduction in cattle holdings. Total losses were estimated at US$1 billion. 14 innovative insurance schemes for agriculture, such as index-based weather insurance products9. Climate change is expected to exacerbate the impact of these extreme weather events. A World Bank study (2013) has shown that changes will lead to variations in temperature and rainfall patterns, and that over the next 40 years climate change will have more severe effects on Moldova. Without decisive adaptation strategies, most yields will suffer significant reduction. Figure 1-9: Effect of Weather on Prices of Crops Figure 1-10: Agriculture GVA Growth and Livestock, 1992–2012 Decomposition, 2002–13 (Current prices, MDL, 2000=100) (Percentage points) 0.4 drought drought drought 0.3 0.2 0.1 0 200220032004200520062007200820092010201120122013 -0.1 -0.2 -0.3 Capital Labour Land TFP -0.4 Source: WDI data Source: WDI data 11. Agriculture products constitute a large share of total exports. Since 2000 Moldova’s agro-food exports have accounted for 60 percent of total merchandise exports. Between 2000 and 2013 the value of agro-food exports went up by 220 percent, to US$935 million. While exports of animal products expanded only moderately, Moldova managed to increase exports of plant-based products seven-fold. As Figure 1-11 shows, specialization in production and exports of vegetal products, especially since 2009, have driven up Moldova’s net exporter position in agro-food products, trade in food products has been balanced, and trade in animal products has been falling. The most widely exported products are both wine and spirits and both fresh and processed fruit and vegetables. These categories alone account for nearly 40 percent of exports. Expanding agricultural production despite Russian restrictions on Moldovan exports has led to shifts in the geographical pattern of exports toward the EU (Figure 1-12). 9 Detailed discussion of the sources of volatility and its effects is presented in “Vulnerability of Moldova’s Agriculture to C limate Change”(2013) 15 Figure 1-11: Value of Moldova's Agro-food Exports Figure 1-12: Destinations for Moldova's Agro-food (Million USD) Exports (Share in total, percent) 1000 1.0 800 0.8 600 0.6 400 0.4 200 0.2 0 0.0 animal (HS01-05) vegetables (HS06-15) food products (HS16-24) EU28 Russia Belarus Ukraine Turkey Other Source: WDI data. Source: WDI data. 12. Trade integration with the EU will generate many opportunities and advantages but will also challenge Moldova’s producers to become more efficient. In 2014 Moldova signed the Association Membership Agreement (AMA) with the EU and the related agreement on the Deep and Comprehensive Free Trade Area (DCFTA). With the opening of the EU-28 economic borders, the Moldovan agro-food sector has a great opportunity to compete in the largest market of the world. This access could bring significant gains to the Moldovan economy through increases in production, revenues, and producer incomes. It could also lower domestic prices via cheaper intermediary inputs entering the Moldovan technological chains. On the other hand, Moldova needs to prepare the agricultural sector to compete with EU producers. Although Moldovan producers have already made major efforts to modernize the sector, significant investment and institutional change are still required to ensure that the agricultural and food industry comply with EU market requirements. 1.2. Policy Perspective 13. Moldova’s strategic vision for agriculture as part of its growth agenda gives priority to sector modernization and improved market integration. Developing the agricultural sector is seen as an essential component of the government’s goal for achieving sustainable economic growth, as reflected in the GoM 2020 National Development Strategy (“Moldova 2020”). The strategy calls for a shift from the current consumption-based growth model towards a growth model based on raising investments, increasing productivity and competitiveness, developing export industries, and promoting a knowledge-based society. Agriculture sector is presently viewed as a potential driver of growth rather than a poverty reduction vehicle; policy measures for in the sector are primarily targeted to supporting market competitiveness and investment-enhancing developments. 14. A sizable budget is required to finance the National Agriculture and Rural Development Strategy for 2014–20 (see Chapter 2). Present policies are guided by a new strategic document adopted in 16 June 2014, the National Agriculture and Rural Development Strategy (NARDS) for 2014–20. The document specifies priorities for the agricultural and food sector through 2020. The budget required to finance the new NARDS is €2.2 billion for the seven-year period (2014–20). Direct financing from the budget will be complemented by large resources from international financial institutions (IFIs) and donors. It is critical that money spent on the sector is aligned with government priorities and sector needs. It is also important to remember that tax preferences are another important form of support to agriculture, which should be directed to sectoral priorities. 15. The challenge of building competitiveness and reorienting a large chunk of Moldova’s agro-food exports toward the EU today seems more pressing than ever in view of the increasing uncertainties about Moldova’s traditional CIS markets. Enhancing Moldova’s agricultural competitiveness is vital for improving the access of Moldovan agro-food products to the EU market and capitalizing on its potential benefits. At the same time, it is realistic to expect that many farmers will continue to produce for non-EU markets and that agriculture will retain its social role as a source of income for less prosperous households. 16. Moldova has much to learn from the recent accession of Central and Eastern European countries to the EU. Review of agricultural developments in New Member States of the EU (NMS) leads to a number of conclusions: (1) Accession had a positive impact on the sector. It resulted in a consolidation of production, higher current prices, higher exports and imports, and especially higher farmer incomes. (2) There is diversity behind the overall positive picture, due to initial conditions, pre-accession policies, post-accession policies, how the Common Agricultural Policy (CAP) was implemented, and the macro policy and institutional environment. The post-accession performance of the NMS was heavily influenced by the agricultural policy prevailing in each country pre-accession. Candidate countries implemented quite different policies, some with positive elements, others with negative experiences after accession and introduction of the CAP (Swinnen- Rozelle, 2009). 17. How countries directed public support and pre-accession facilities provided by the EU influenced how agriculture performed after accession. The most critical component of pre-accession policies was not the level of support to agricultural production but its purpose and how countries used pre- accession EU-provided facilities. Public support was beneficial post-accession if in the pre-accession period it had been mainly concerned with enhancing long-term competitiveness. In Romania, Hungary, and the Czech Republic, the level of support was close to or even higher than the EU average but it did not bring significant positive results (e.g., price increase) because it was too focused on price and market support. Countries giving priority to enhancing competitiveness, modernizing the agri-food industry, and adapting it to EU standards benefited more in terms of how agriculture performed post-accession; Poland is a good example (see Annex 1). Restrictive land policies, as in Hungary, and failure to consolidate land negatively influenced utilization of the advantages of larger markets by constraining significantly the flow of outside capital to agricultural (Csáki et al. 2008). In contrast, liberal land policies, as in the Baltic countries, helped the agricultural sector to obtain more resources and better take advantage of the possibilities created by accession and close integration with the EU. 17 1.3. Moldova’s Fiscal Challenges and Public Support to Agriculture 18. The Government of Moldova managed to maintain fiscal discipline in 2010–14, but it does not have fiscal space for a substantial increase in spending. After undertaking the adjustment program that was supported by the IMF, after 2009 the general government deficit declined by almost 4 pp of GDP, to 1.8 percent in 2013–14. Moldova contained the increase in public wages, and made spending on education and social assistance more efficient (see Figure 1-3) However, budget amendments introduced in summer 2014, including wage and pension raises, lifted recurrent expenditures by 0.6 pp of GDP in 2014 and another 1 pp in 2015. The government foresees a 1 percent contraction of economic activity in 2015 and adopted a budget that targeted general government deficit of 3.8 percent of GDP. Interest expenditures are also going to increase significantly in the medium term once the state recognizes the losses incurred in the banking sector. Thus, debt sustainability considerations do not allow Moldova to spend more without jeopardizing macroeconomic stability. 19. Government spending is already high for Moldova’s level of income; the path forward lies through better prioritization and more efficient use of public resources. Since early 2000 public spending has risen steeply, from about 26 percent of GDP in 2001–02 to over 38 percent in 2014, even after a considerable decline from the crisis-related peak of over 42 percent. The spending increase was driven by an expansion of capital spending, which jumped to 8 percent of GDP, supplemented by higher spending on wages and pensions as well as goods and services (Figure 1.13). From the functional side, Moldova spends more on economic affairs, health, and education than comparators. As a result, Moldova’s spending exceeds that of other countries with similar or even higher per capita income (Figure 1-14). This calls for revisiting the role and optimal size of government, better prioritization, and more efficient use of public resources. 20. Taxes are relatively high, which limits room to maneuver on the revenue side. Moldova’s tax revenue is close to the average for its peers but higher than in some more developed CEE countries like Poland and Slovakia (Figure 1-15 and 1.16). The tax structure is somewhat skewed toward consumption taxes, which is a positive, growth-friendly feature (Arnold 2009). In 2014 total revenues and grants were 39 percent of GDP and tax revenues equal to 32 percent of GDP. Since 2008 tax revenues have been relatively stable but the structure has changed: a decline in revenues from value-added tax has been offset by an increase in CIT tax. As result, about 30 percent of taxes collected are now from VAT and excises, followed by SSCs of about 20 percent. The share of income taxes in government revenues is minimal: less than 5 percent of GDP. The level and the structure of taxes in Moldova do not leave much space for more taxation to fund higher spending. 18 Figure 1-13: Government Spending Structure, Figure 1-14: Government Spending vs. GDP per Moldova, Capita, 2012 (Percent of GDP) (in percent of GDP and US dollars, PPP) Source: Ministry of Finance data. Source: WDI and WB staff calculations. Figure 1-15: Tax Revenue Structure, Moldova, Figure 1-16: Tax Revenue Structure, 2012 (Percent of GDP) (Percent of GDP) 40 30 20 10 0 GEO BGR ROM MKD MDA BLR UKR SRD Other Taxes SSC Taxes on Goods and Services Taxes on International Trade Taxes on Income and Profits Source: Ministry of Finance. Source: IMF GFS. 21. In the medium term, Moldova’s fiscal sustainability will be challenged by the institutional and social context. On the spending side, social and demographic challenges will intensify, building on current social insurance and pension commitments. Aging will push up the pension bill and fuel demand for health care services, while public spending on education and health in particular remains high. Intergovernmental fiscal relations should be reformed in a way that is consistent with fiscal discipline and reduces the inefficiencies of small, fragmented local governments. Further spending adjustments are needed to make space for pressing medium-term public investment needs. On the revenue side, institutional weaknesses are reflected in tax collection problems. The scope for raising revenues is limited because tax revenue is already high as a percent of GDP, compared with regional neighbors. There are limited possibilities for raising revenues in real estate taxes, VAT—not only (because it is the largest revenue source, but because there are numerous exemptions with zero or reduced rates—and excise taxes. Nevertheless, the main (but long-term) 19 possibilities for raising revenues should not be from raising taxes, which can reinforce compliance problems, but from improving collection and therefore administration. Moreover, Moldova could see a decline in external assistance, making it important to preserve fiscal sustainability in the medium term. 22. Moldova spends more on agriculture than other countries at the same level of development. Its agricultural spending at about 1.4 percent of GDP is higher than that of most benchmark countries, including such new EU member states Figure 1-17: Government Spending on Agriculture as Romania and Bulgaria.10 Only (Percent of GDP) Belarus and Uzbekistan spend more (Figure 1-17). Moldova’s actual agricultural spending reflects a surge in recent outlays on the sector: between 2002 and 2014 spending on agriculture as a share of GDP tripled. Only Georgia had a larger rise, but to a lesser extent such spending also jumped in Bulgaria and Croatia. During that period the other comparator countries reduced spending. At 3.6 percent of total Note: Including Forestry and Fishing. government spending, Moldova’s Source: Data from BOOST, FAO, IMF. public spending on agriculture compares favorably to that of many ECA countries (being outperformed only by Uzbekistan, 5.4 percent of total spending, and Belarus, 6.8 percent of the total). Chapter 2 provides a detailed discussion of spending on agriculture. Public support to agriculture in Moldova is in fact much higher than headline figures show. Tax expenditures add to direct outlays. As Chapter 3 of this report shows, foregone revenues due to tax expenditures exceed 0.6 percent of GDP. 23. Despite the increased resources devoted to agriculture, its productivity has not gone up much. Though close, Moldova is somewhat below the efficiency frontier that represents maximum attainable productivity for a given level of spending (see Annex 2) for an explanation of the methodology). This means that some countries that spend similar amounts achieve better results— higher agricultural productivity. Moreover, Moldova has recently been drifting away from the efficiency frontier (Figure 1.18 and 1.19), which suggests that higher spending has meant higher productivity, at least in relative terms. The same is true for Georgia and Bulgaria, which increased spending between 2002 and 2012. The analysis suggests that Moldova’s spending on agriculture could be made more efficient, either by spending less or ensuring that spending has a positive 10 For international comparability purposes, the Classification of the Functions of Government - COFOG definition labelled 11 in the Boost data is used here. It is different from the local mapping (see Annex 3) used in this report to assess the size of the sector budget. 20 effect on agricultural productivity. The latter could be done by enhancing the quality of spending, perhaps by changing its structure as well as managing is better. Moldova, 2013 Public support Tax expenditures (0.63% GDP): Direct budget spending (1.4% GDP): 1. Reduced CIT rate 1. Subsidies 2. Reduced VAT rate 2. Services 3. Reduced SSC rate 3. Donor programs 4. Reduced health contribution rate Source: BOOST database, MOF, WB staff calculations. Figure 1-18: Agriculture: Government Spending Figure 1-19: Agriculture: Government Spending on Agriculture and Total Factor Productivity on Agriculture and Total Factor Productivity (Free Disposal Hull Method) (Batesse and Coelli (1988) Method) Source: WDI, WB staff calculations. 1.4. Conclusions 24. Despite positive developments in recent years, ensuring further sustainable growth for agriculture will be quite difficult given the less favorable external conditions. Large increases in agricultural output leading to strong performance of agriculture exports, as well as poverty reduction in rural areas, are the main sector achievements in recent years. Over that time, agriculture has proved its resilience and development potential. To sustain the pace of progress, 21 structural improvements are needed, in terms of both investments and innovation to make the sector more competitive and more resilient when external conditions are not that favorable. The next chapter discusses the role of fiscal policy. 22 Chapter 2. Level and Composition of Government Spending on Agriculture Public spending on agriculture in Moldova has gone up considerably, driven by donor-funded projects. Meanwhile, the country has made progress in improving the budget composition to align it with government priorities for the sector. The structure of the budget reflects the three main pillars of the National Strategy for agriculture. In particular, a large and increasing share of capital spending has helped upgrade obsolete infrastructure and modernize the sector while increasing the sustainability of natural resources by putting more effort into risk management. The aspiration to improve conditions for living and working in rural areas cannot be attributed to the sector budget only. It involves several departments and spans areas much broader than what are usually covered by agriculture. However, not all types of intervention financed from the agriculture budget are efficient. In general, the substance and the scope of the recurrent subsidy instruments now in place are not the most efficient responses to sector constraints. For instance, subsidies to facilitate access to credit are primarily benefiting wealthier farmers, who are already in a position to qualify for commercial credit. This support measure is also used mainly for subsidizing operational loans, so it has become a vehicle for subsidizing agricultural inputs with only a small share used for investments to build productive capacity (such as buying equipment and rehabilitating infrastructure). Such public services as agricultural education, research, and irrigation are inefficient and need major restructuring. There are also design issues related to risk management programs, such as hail prevention and the insurance subsidy program. Finally, public support for agricultural investment is captured by a small number of corporate farms and often does not reach individual small farmers, who are the poorest. It is also affected by gender and geographical inequalities. This chapter of the Agriculture Public Expenditure Review assesses whether agricultural spending supports sector objectives. It analyzes the nature, composition, and efficiency of public spending on agriculture in order to identify what must be done to make the budget a relevant tool for supporting sector objectives. The rest of this chapter is organized as follows: the first section discusses trends in agricultural spending. The second looks at the efficiency of public spending in order to determine whether scarce public resources are allocated in ways that reflect government priorities. The third section analyzes the effectiveness of agricultural programs by tracking their results and impact. The fourth section looks at the distributional impact and the inequality aspects of spending on agriculture. The concluding section summarizes the major findings and policy recommendations. 2.1. Trends in the Budget for Agriculture 25. Government spending on agriculture has gone up considerably since 2002, though not as much as total spending. Outlays on agriculture rose from less than 0.5 percent in GDP in 2002 to more than 1.5 percent in 2014 (Figure 2-1). After a post-crisis decline in 2010–11, agricultural spending 23 reversed and by 2014 it exceeded the 2009 level, due mainly to one-off compensation to farmers to offset Russian trade restrictions and to an inflow of donor funding. Excluding tax expenditures, government support to agriculture amounted on average to about 2.6 percent of the total budget for 2009–13.11 Spending on agriculture as a share of spending has been stable unlike the declining trend (as share of the total government expenditure) observed in other developing countries after the crisis in 2008-09.12 Figure 2-1: Moldova: Public Spending and Figure 2-2: Agriculture Budget, 2009–13, Spending on Agriculture, 2002–14 (Percent of GDP) (Percent of GDP) Source: National authorities. Source: BOOST data. Table 2.1: Structure of General Government Expenditure by Function, 2009–13 (Percent) Sectors 2009 2010 2011 2012 2013 Public order, security and justice 14.0 11.6 12.1 12.0 12.7 Education, research and innovations 18.5 21.3 21.1 20.9 18.8 Health 14.0 13.5 13.1 12.9 13.4 Social care and social insurance 32.2 34.4 33.9 32.8 33.3 Agriculture 2.6 2.7 2.5 2.5 3.1 Industry, construction, transport, etc. 7.6 8.1 8.8 10.0 10.0 Other outlays 11.1 8.5 8.4 8.9 8.7 Source: BOOST data. 26. The increase in public spending on agriculture was mainly due to an increase in donor funds for investment. External resources, predominantly investment loans and grants from multi- and 11 The sector budget includes the Ministry of Agriculture, the budget for the irrigation unit at the Ministry of Environment, and the budget of the Ministry of Education and Academia for Research. See Annex 3. 12 Most developing countries seem to have prioritized other sectors in response to the crisis. Africa experienced a substantial decline in the agriculture share of government spending, from an average of 3.5 percent in 2009 to about 2.5 percent in 2011; Asia saw a contraction of 0.6 pp., from 4.4 percent in 2009 to an average of 3.8 percent in 2011. 24 bilateral partners, rose by 0.4 percent of GDP13 from 2010 to 2013. In general, international support for Moldova has been growing since 2006, with about one-third of the funds directed to agriculture. These funds were allocated to high-priority investment to modernize the sector (Figure 2-2) 2.2. Key Aspects of Economic and Functional Composition 27. The share of capital outlays in the agriculture budget rose at the expense of recurrent expenditures— a welcome trend that sets Moldova apart from other ECA countries. From about 0.6 percent of GDP, capital outlays now account for nearly 0.8 percent and recurrent spending has plunged (Table 2.2). The change is seen as a positive development that is addressing investment and modernization needs and helping to reduce inefficient recurrent subsidies. Moldova’s emphasis on capital spending contrasts with other countries in ECA: in FYR Macedonia, nearly 98 percent of agricultural spending in 2013 was on recurrent items and in Ukraine it was 91 percent. Table 2.2: Recurrent vs Capital Agriculture Spending 2009 2010 2011 2012 2013 Percent of GDP Capital 0.68 0.57 0.53 0.93 0.85 Recurrent 1.27 0.87 0.65 0.65 0.57 Percent of Total Budget Capital 35.1 39.4 45.1 58.8 59.8 Recurrent 65 60.6 54.9 41.2 40.2 Source. WB staff based on BOOST. 28. The structure of recurrent spending has also improved, with outlays shifting away from subsidies. While goods and services have held steady and wages have been controlled, transfers (recurrent subsidies) have dropped from 0.7 percent of GDP to 0.1 percent. The wage bill accounts for 40 percent of the budget. Table 2.3: Economic Structure of Recurrent Spending (Percent of GDP) 2009 2010 2011 2012 2013 Wages 0.40 0.30 0.30 0.30 0.20 Transfers 0.70 0.40 0.20 0.20 0.10 Goods and Services 0.20 0.20 0.20 0.20 0.20 Source. BOOST data 13 According to BOOST data. 25 29. The functional composition of the budget shows a concentration of spending on a few items that are critical for the development of agriculture. The largest shares of agriculture spending are allocated to physical infrastructure and business development to modernize the sector, and to key services, the largest shares being for food safety and agricultural education, with smaller shares for research, extension services, and irrigation. These are followed by viticulture and wine development in support of high-value markets and risk mitigation and anti-hail programs to help make agriculture more resilient to weather risks. As shown in Table 2.4, the share of these items increased from 41 percent in 2009 to almost 84 percent in 2009-2013, reflecting a concentration of public resources on these programs critical for the modernization agriculture and its sustainability. Table 2.4: Functional Classification of Executed Budgets, 2009–13 (Percent of total spending) 2009 2010 2011 2012 2013 Functional breakdown, executed Agricultural education 10.12 12.15 13.26 10.23 9.96 Research 5.57 6.95 8.23 5.42 5.41 Food Safety 10.03 10.92 11.55 9.90 9.67 Anti-hail Service 3.57 4.25 5.29 3.95 5.61 Irrigation 3.16 2.76 2.42 1.46 1.62 Seed capacity development 0.54 0.72 0.80 0.71 0.80 Advisory Service 0.00 0.00 0.00 0.00 0.98 Animal Identification & Traceability 0.65 0.90 1.04 0.67 0.47 MAFI, general administration 0.71 0.84 1.13 0.99 1.02 Other services, MAFI 0.43 0.66 1.33 1.08 0.85 Agriculture risk mitigation 2.24 4.45 3.84 9.48 6.05 Viticulture and wine development 4.37 7.59 3.71 13.66 12.17 Support to farmers (Subsidies non id) 29.94 16.77 9.63 2.92 3.39 Support to tree industry development 6.46 3.68 4.53 3.13 4.01 Support to tobaco industry development 0.00 0.00 0.00 0.00 0.00 Support to animal husbandry 1.82 1.18 2.71 2.01 3.98 Physical infrastructure and business development 13.37 21.38 27.63 32.32 26.59 Other services (Donors) 6.89 4.23 2.33 1.81 6.77 Other services (Local) 0.13 0.58 0.55 0.25 0.65 Total sector 100.00 100.00 100 100.00 100.00 Source: WB based on BOOST data. 30. Services account for more than 30 percent of the agricultural budget. The core public agricultural services are operational. The annual funding allocated to agricultural services mainly covers the recurrent needs of the institutions, with no capital resources for their effective functioning and institutional development in general. This holds true for large budget items like education, food safety, and research, as well as smaller services like irrigation. Some agricultural institutions benefit more than others from additional annual allocations from donors, which help fill critical gaps (in both know-how and investment) not covered by the public budget. Favorite areas for donor support are food safety, irrigation, and advisory services. Such key services as research and education are primarily state-funded and have so far seen limited donor support, possibly because of pending government-led structural reforms in these areas. Annex 4 provides a summary of agricultural services along with problems specific to each. 26 27 Table 2.5: Functional Classification of Executed Budgets for Services and Subsidies, 2009–13 (in thousands of MDL and percent) Agriculture Budget 2009 2011 2013 2013 MDL thousands Percent of Total SERVICES (MAFI) 307 334 418 29.6 Of which: Agricultural education 119 129 142 10.0 Food safety 118 112 138 9.7 Anti-hail service 42 51 80 5.6 Testing & registration of new crop varieties 6 8 11 0.8 Animal Identification & Traceability 8 10 7 0.5 MAFI central office 8 11 15 1.0 Other MAFI services 5 13 12 0.9 SERVICES (Other ministries) 87 101 98 6.9 Agricultural research (Academy of Sciences) 66 80 77 5.4 Irrigation (MinEnv) 22 22 21 1.5 SUBSIDIES (MAFI) 612 404 463 32.8 Subsidies 611 401 463 32.7 Emergency support 1 3 0 0.0 DONOR PROJECTS 171 124 434 30.7 TOTAL 1,177 963 1,413 100.0 Source: Boost Data – World Bank, 2013. 2.3. Effectiveness of Public Spending in Supporting the National Sector Strategy 31. The structure of spending on agriculture in Moldova is aligned with the priorities defined in the National Agriculture and Rural Development Strategy (NARDS). Spending is guided by the main pillars for the sector: (1) enhanced competitiveness of the agricultural and agri-food sector through restructuring and modernization; (2) sustainable management of natural resources; and (3) improved conditions for living and working in agricultural and rural areas. More than 80 percent of agriculture spending supports the first priority, competitiveness, mainly by investments to modernize the sector and delivery of agricultural services, such as enhancement of food safety and agricultural education. About 15 percent of agricultural funds are spent to ensure sustainable management of natural resources. Vulnerability to climate distress is addressed by weather-related risk mitigation instruments (access to irrigation, adoption of modern agricultural technologies such as drought-resistant varieties, anti-hail protection) and innovative insurance schemes (such as index-based weather insurance). Realization of the government’s third priority, improved conditions for living and working, is more influenced by donor projects that provide funding for activities contributing to this objective. Funding of this priority also spans areas beyond the responsibility of the Ministry of Agriculture and involves other ministries. 28 Table 2.6: Sector Priorities and Allocation of Agriculture Spending Percent of Total Priority Spending Category Amount (MLD) Spending Pillar I. Enhanced competitiveness through restructuring and modernization 1,140,712,526 82.8 Investment subsidies: machinery and equipment, livestock equipment and breeds, post-harvest and processing equipment (M3, M5, M6, M7, 343,982,861 25.0 M8) Component A: Increased investments for modernization Donor- financed projects (WB, IFAD, MCC, EIB) 369,237,736 26.8 Component B: Improved access to capital and input and Current subsidies: credit and land consolidation (M1, M9) 39,341,478 2.9 output markets for farmers Component C: Enhanced agricultural education, Services : food safety, agricultural education, research, extension, testing research, and extension services and strengthened 388,150,451 28.2 and registration of new crop varieties, animal identification synergies between them; a reformed food safety system Pillar 2. Sustainable management of natural resources 171,291,890 12.4 Current subsidies: risk insurance, energy for irrigation (M2, M10) Component A: Reduced climate-related risks for Capital subsidies: irrigation equipment, anti-hail and anti-frost systems, 70,631,072 5.1 agricultural production protected field vegetable production (part of M3, M5, the whole M4) Services: irrigation and anti-hail 100,660,818 7.3 Pillar 3. Improved conditions for living and working in agricultural and rural areas 65,159,600 4.7 Donor- financed projects 65,159,600 4.7 29 32. The rest of this section assesses the efficiency of the main programs supporting Pillars 1 and 2 of the sector strategy, to which 95 percent of sector spending is allocated. This approach does not differentiate between the nature of support according to its economic (investment or recurrent spending) or functional composition but rather focuses on the results achieved. Pillar I. Enhanced competitiveness through restructuring and modernization 33. Donor programs helped to reorient spending toward activities supporting competitiveness and modernization. Donor spending on the wine-growing industry helped redirect domestic spending toward investments in machinery and equipment. Apart from supporting the wine industry, international and donor programs finance other areas of the government’s agricultural competitiveness agenda: modernization of the agri-food chain, food safety and quality requirements, sustainable land management, development of the horticulture value chain, and support for dealing with resource endowment and depletion. As a result, the government has redirected its capital spending toward machinery and equipment, especially for improving the organization of production and post-harvest collection to facilitate selection, packaging, and contracting procedures in line with European practices. 34. Moldova could reap the benefits of higher donor funding in many ways: (1) The bulk of the funds come from the EU, which recently signed the DCFTA cooperation agreement with Moldova. This could provide some stability for future funding. (2) All large donor projects are presently captured by the public budget; what is not yet captured are technical assistance projects (mainly EU/FAO, USAID), which account for less than 10 percent of total donor support for agriculture. (3) Donor- supported investments are closely coordinated with the GoM, including the annual Agricultural Support Fund instruments, and some donor funds to the sector are channeled through MAFI’s Agency of Interventions and Payment for Agriculture (AIPA). Thus, donor funding complements the national budget positively. 35. Investment subsidies support the introduction into agriculture of modern and innovative production and post-production practices, with emphasis on high-potential high-value agriculture (HVA). Subsidies account for about 40 percent of investment spending in Moldova— over 90 percent if donor funding is excluded. The bulk of investment support to crops is channeled to the horticulture value chain to support productivity and market competitiveness improvements; equipment for post-harvest handling and processing of fruit and vegetables, and planting of new orchards and vineyards. All investment subsidies represent grants to producers that cover a portion (normally 25–50 percent) of their investment in productive assets. 30 Figure 2-3: Structure of Domestically Funded Figure 2-4: Distribution of Investment Investment in Agriculture, 2009–13, Subsidies, 2012–1414, (Percent) (Million USD) 1.0 16 Other capital 2012 2013 2014 0.9 transfers 14 0.8 Purchase equipment 12 0.7 and appliances 10.1 10 0.6 Support animal 8.4 husbandry 8 6.7 0.5 0.4 Support fruit trees 6 0.3 industry 3.6 3.4 3.4 4 0.2 Support wine- 2 0.1 growing 0 0.0 Purchase of fixed M3 M4 M5 M6 M7 M8 2009 2010 2011 2012 2013 assets Source: BOOST data. Source: AIPA datasets 2012–14. 36. Reducing the concentration of the investment subsidy program and refocusing funding toward “smart”15 subsidies helped to improve the allocative efficiency of subsidies (Figure 2-4).16 The largest machinery program (M5) has been both redesigned and reduced to better respond to sector modernization needs. Some large-scale traditional-type agricultural machinery has been removed from the list of subsidized items and replaced by more innovative types. The currently largest post-harvest infrastructure program (M8) addresses a major constraint of Moldova’s production sector: low value-added due to insufficient capacity for post-harvest handling and processing of primary agricultural products. Demand for the smaller programs (M4 – protected field horticulture production, M6 and M7 – livestock breeding material and technological upgrading) seems to have been picking up over the last two years, indicating more interest in investing in these sectors. 37. Recurrent subsidies intended at building competitiveness may need to be redesigned, though they are not very expensive. Half of recurrent subsidies (only 1.5 percent of total spending) aims to improve the competiveness of agriculture, especially programs supporting access to capital and land. Spending on interest rate subsidies (M1), which focuses on increasing the access of farmers to credit has risen recently and is now the largest of the recurrent subsidies. However, credit is allocated primarily to wealthier farmers, who are already in a position to qualify for commercial credit. Another serious concern is that it is used mainly to subsidize loans for operations, so that it becomes in effect a vehicle for subsidizing agricultural inputs. The 2014 AIPA data reveals that 1414 M3: Anti-hail and anti-frost programs; M4 program for protected-field plant production; M5: programs for machinery; M6: programs for livestock breeding material; M7: programs for technological upgrading; M8: programs for post-harvest handling and processing of primary agricultural products 15 Subsidies stimulating adoption of innovative technologies and approaches are referred to as smart subsidies. 16 See Annex 2. 31 almost two-thirds of subsidized credits went for fuel (25 percent), fertilizers (22 percent), and seeds and seedlings (17 percent); only a small share went into investments to strengthen productive capacity, such as purchase of equipment and work on infrastructure. Figure 2-5 : Distribution of Recurrent Subsidies, 2012– 38. The subsidy program for 14 land consolidation has not (Millions of US dollars) brought results. To address the severe land fragmentation, 5 2012 2013 which is efficient and lowers production, the GoM offers 4 subsidized fees for merging land parcels. The land consolidation 3 program (M9) covers up to 50 percent of the taxes, notary 2 fees, and cadaster costs 1 associated with the consolidation of agricultural 0 parcels. The measure is now M1 M2 M9 M10 capped, in terms both of the subsidy per ha or transaction Source: AIPA data 2012–14. and the eligible number of hectares. This facilitates more even distribution of subsidies among farmers, but its current uptake is very limited at US$8,000– 9,000 annually, with only nine applicants in 2014 and six in 2013. As a result only about 1,000 land parcels were consolidated. The low uptake of this program reflects the current slow pace of valuation of land markets although the pace is expected to pick up in the near future. Although there is a power rationale for it, if scale and critical mass are not achieved in the next few years, administrative costs would most likely not justify retaining this measure. 39. Services supporting Pillar I have been sizable and stable, with about 30 percent of agricultural spending allocated to research, education, food and safety, and extension services. According to international studies, these services can be a strong return to public investments and boost agriculture productivity and competitiveness. While there are efforts at improving the quality of food safety, the effectiveness of some other services is questionable due to deep-rooted structural inefficiencies built into the current systems. Most of the problems identified are complex and require wide-ranging approaches and in some cases bold structural reforms (see Annex 4). Services that deliver results have only a small share in the budget. 40. Research and education, which are large budget items, offer limited value-added due to outdated methods and systemic deficiencies: (1) Research and education are driven by a sizable wage bill that results from a heavy institutional structure. The wage bill for agricultural education and research institutions in Moldova accounts for about 10 percent of total agricultural spending. 32 Because the funding is spread over several institutions, it has little ability to generate outcomes, such as a higher (quality of education or research spillovers to extension services). (2) The link between research and the needs of farmers is missing. The research institutes operate in isolation from the private sector and from the international R&D system. The current structure of agricultural education does not respond to the current needs of the country and does not generate positive results (see World Bank, 2007a). The curriculum in most colleges is outdated and not responsive to the requirements of a modern agricultural industry; nor is there a useful link to the private sector. For instance, it does not provide on-farm training to introduce the certification demanded EU retailers demand or to meet Global G.A.P. (Good Agricultural Practice) and other private standards, which is a good way of establishing closer links with the private sector. 41. Coverage and efficiency of the extension network are high and have potential for improvement despite the network’s small share of the sector budget. Presently, the network covers over 40 percent of Moldova’s territory and over 50 percent of farms. Most of the advice offered relates to production technologies (52 percent of services in 2014); other consultancy areas are marketing (16 percent of services), business (20 percent), and legal advice (12 percent). To complement and maximize the efficiency of extension services, the role and operation of Seed Control (the State Commission for Testing Plant Varieties17), which is mandated by Moldova’s legal requirement is widely discussed in Moldova. The Seeds Commission would function better if its role were to shift from a prescriptive to a consultative one by abolishing the mandatory requirement to test new varieties that are used in other countries although not yet registered in Moldova. This change would better serve the needs of the private sector by offering immediate access to new plant varieties developed elsewhere. 42. Modernization of the food safety system will be continued, but public funding in this area could be further optimized. The spending on food safety absorbs about 10 percent of total spending on agriculture and is allocated to ANSA which is in charge of ensuring control over the entire food chain. ANSA, with strong support from donors, is now proactively engaged in harmonizing its legal and institutional framework with EU practices and requirements. Though significant progress has been made in various areas of food safety and quality control in Moldova, a robust, reliable and effective domestic food safety system has yet to be built. Among the elements needed are (1) aligning food safety requirements and norms with internationally accepted practices; (2) optimizing the ANSA laboratory to make lab testing more reliable and achieve laboratory accreditation; (3) transitioning to risk-based inspections and controls; and (4) setting up advanced information and data management systems. Pillar II. Sustainable management of resources 43. Risk management programs to facilitate sustainable management of resources account for 15 percent of total spending. Budgeted agricultural risk management programs are mainly delivered 17 Any new variety must be tested before it can be sold on the domestic market. The clients of the Seeds Commission are either local breeders (mainly agricultural research institutes) or seed importers that want to introduce a variety that is not yet registered in the national catalogue of plant varieties. There is a charge. 33 by government-funded services investment subsidies (both recurrent and investment, Table 2.1). The main programs are shown in Table 2.7. Table 2.7: Risk Management Package (RMP) 2014 (Millions of US dollars) Subsidies Investment M3 anti-hail and anti-frost systems 0.22 M4 protected-field plant production 3.63 M5 irrigation equipment 2.50 Subsidies Recurrent M2 agricultural insurance 2.09 M10 energy for irrigation 0.00 Services Anti-hail Service (MAFI) 5.68 Water Service Agency (MEnv) 1.49 TOTAL 15.61 Source: BOOST and AIPA data.  Services. The government has for many years operated a hail prevention service using an expensive system of ground-based rockets and radar stations. The Water Service Agency (Apele Moldovei) that manages the large-scale state irrigation systems also requires an annual public budget of about $1.5m (part of the Ministry of Environment budget because the water agency is subordinated to this ministry).  Subsidies. A number of investment subsidies are targeted to building up agriculture’s resilience to weather risks, such as support to purchase anti-hail and anti-frost systems (part of M3), irrigation equipment (part of M5), protected field vegetable and strawberry production (M4). Two recurrent programs, – agricultural insurance (M2) and energy for irrigation (M10), are also part of the RMP. The following section assesses the efficiency of these programs. 44. The cost-effectiveness of a hail prevention system is questionable. This service has been absorbing increasing amounts of budget funds every year; its share in the total services budget has gone up from 10.3 percent in 2009 to 15.5 percent in 2013. However, international research has found no scientifically credible evidence worldwide that hail can be suppressed, so the benefits of a rocket-based anti-hail system are unproven. A cost-benefit analysis to compare the efficiency of the rocket approach to a risk-reduction approach based on hail insurance would be informative and useful (World Bank 2007). Similarly, there is little evidence that the traditional insurance system, like the one in Moldova, is well- suited for managing national systemic risks, such as drought, so it is an inadvisable option for Moldova (World Bank, 2007). 45. The irrigation system is underutilized and its management is not sustainable. The bulk of annual budget allocations for irrigation (about 75 percent) are staff costs. The rest is for maintenance and rehabilitation of irrigation systems. It is unlikely that this amount is adequate for proper 34 maintenance. As a result, the Water Agency, which is responsible for the irrigation system,18 performs below its potential, and only 7,000–15,000 ha of the 144,000 ha of agricultural land are irrigated every year (out of 121 operating irrigation stations, only 40–50 are used utilized). Earlier studies on this topic recommended transferring irrigation management to Water User Associations as the only viable option for long-term sustainability of Moldova’s irrigation systems. This is being addressed as part of the MCC irrigation program involved in rehabilitating 10 central irrigation systems (to be finalized in 2015), but these efforts need to continue and public funding needs to be scaled up. 46. The irrigation subsidy has not yet been successful in attracting sizable investments in small-scale on-farm irrigation. Despite an extremely high rate of re-pumping of irrigation water (80 percent subsidy rate), which is seen as “unhealthy” and creating an extremely high dependence on the subsidy, the 2013–14 uptake of the measure was still low ($0.3m in 2014), with only about 40 annual beneficiaries irrigating a total of 4,500–5,000 hectares of land. The main reasons for this dismal performance are (1) approval of the subsidy is conditioned on proof of having paid water fees, which accounts for about 60 percent of the subsidy, which may be why the low subsidy is not attractive for some farmers, and (2) most claims for the subsidy comes later in the year when the subsidy fund is nearly depleted, so the amount approved is either subject to a coefficient application or in some cases held for payment in the next calendar year. 47. The risk insurance subsidy program (M2) has been in place since 2004 but does not appear to have had encouraging results in stimulating faster adoption of agricultural risk insurance. The amounts annually allocated to this measure had been Table 2.8: Main Risks Insured in Moldova, 2011–2014 gradually growing from (Percent) $1.2m in 2011 to $2.8m in Insured risk 2011 2013 2014 2012 and to $3.3m in 2013. Hail 38 43 39 Yet in 2011–total insured Drought 23 38 34 area held steady at about Frost 2 18 17 20,000–30,000 hectares, Animal diseases 1 1 10 about 1–1.5 percent of total Source: AIPA arable land. The main risks insured are hail and drought (Table 2.8) The scheme has benefitted only a handful of agricultural producers (up to 200 beneficiaries a year), and the number of annual insurance contracts seems to have been declining in the last few years,19 even though annual allocations to this measure have been going up. There are also anecdotal reports of insurance companies using fraudulent practices by supplying fake documentation to get higher subsidy payments.20 18 Expenditures for irrigation are allocated to the Water Agency (subordinated to the Ministry of Environment) that is in charge of managing the large-scale central irrigation systems. The Water Agency manages 78 irrigation systems throughout the country. 19 Number of subsidized insurance contracts: 107 in 2014; 78 in 2013; 266 in 2012; 170 in 2011. 20 The subsidy is claimed by the insurance company (50 percent of the premium set in the contract), not the farmer-beneficiary. 35 2.4. Distributional Impact and Inequality of Agricultural Subsidies 48. Public spending on agriculture seems to reach primarily large farms, creating inequalities that run counter to the inclusiveness intent of Pillar 3.21 Inequalities of access to public spending reflect the size and legal status of farms, and geographic and gender dimensions.22 Today, the overwhelming majority of subsidies are absorbed by large corporate farms—75–80 percent for the past three years. About 71 percent of farms that received any subsidy in 2010 had more than 100 ha of land, and 98 percent had more than 5 ha. The only subsidy largely benefitting smaller individual farms23 is the one stimulating investments in protected-field vegetable production. These outcomes are not consistent with the weight of individual farms in total agricultural output, which is significantly higher than that of the corporate sector: 70 versus 30 percent. 49. Small farmers face greater information and monetary constraints than large farmers in accessing modern technology. Improving access to better services should raise yields and incomes for small farmers. Ensuring that expansion of services Figure 2-6: Corporate and Individual Subsidy programs is targeted at smaller farm units Beneficiaries, 2012 would help to make these subsidies more (Percent) progressive. Similarly, reducing the share of resources absorbed by the largest machinery PH Infrastructure program would lead to a more equitable and Liv Breeds efficient distribution of agricultural Liv Equipment investment subsidies. Machinery 50. While it is common that an Vegs investment-led subsidy program benefits Vineyards first and foremost farmers with the most Orchards resources, it is important to keep the focus Insurance on expanding program coverage to Credits prospective smaller-scale farmers. 0% 20% 40% 60% 80% Investment subsidies are by definition more 100% corporate sector individual sector likely to be captured by most resourceful farmers. However, limiting, or even Source : AIPA , abolishing, subsidies for traditional large- *Recurrent subsidy. scale agricultural machinery and equipment (primarily captured by large farms) and refocusing support on high-tech technologies with an 21 A comprehensive analysis of the distributional implications of public spending on agriculture is complicated by the lack of data on both small and commercial farms and information on the welfare of farm owners. 22 Poverty rates are highest among farmers (36 percent) and agricultural workers (45 percent); together, the two categories account for 40 percent of Moldova's poor. Data from the 2012 Household Budget Survey suggests that 44 percent of those in the bottom 40 percent of Moldova’s welfare distribution are farmers or farm workers. 23 The available data does distinguish between corporate and individual farms, although specific farm sizes are not provided. It is important to realize that the size of individual farms can vary, so that “individual farms” are not always “small farms” (and in fact can be fairly large enterprises). However, per unit data clearly suggests that individual farms are in general smaller businesses than the corporate enterprises. 36 emphasis on the HVA sector (where most operators are small and medium farmers), can be expected to lead to a more equitable distribution of public capital resources. Capping investment subsidies—in tandem with setting differentiated subsidy rates that taper off—will serve the same purpose by allowing more farmers to participate in subsidy schemes, and reduce the gap in the subsidy amounts obtained by large versus small farms. 51. There are also regional disparities in public spending on agriculture. The South is the poorest region in the country but farmers there received fewer subsidies than those in the North or Central regions Subsidy reallocation also specifically benefits individual farmers in the South, but they receive even lower subsidies than the average. Redistribution of subsidies should have a deeper impact on regions where the concentration of poor is highest but allocation of subsidies was previously lowest, but such redistribution needs to be facilitated by concerted efforts and be based on objective criteria that track sector objectives. 52. Finally, it seems that gender is another dimension of disparity in the allocation of subsidies. For both individual and corporate farms, the share of subsidy recipients is slightly lower among female- headed than male-headed farms: 27.1 percent of all female-run corporate farms receiving a subsidy and 30.2 percent of male-run farms); and 6.1 percent of all female-run individual farms receive a subsidy compared to 7.4 percent for male-run farms.24 Increasing inclusion of small farms in the agriculture support program would benefit especially female farmers, since 36 percent of individual farms have a female head, though only 14 percent of corporate farms. 53. The distributional implications of supporting large-scale agriculture as opposed to small scale farming cannot be dealt with in depth here. The approach adopted by the MAFI to support the sector is the following: within a limited budget envelope, the objective is to support market- oriented and competitive farmers who will contribute to sustainable sector growth and development. To the extent that the government wants to support large-scale farming, it would be important that the decision be based on economic arguments, such as the need to address some market failure, rather than being a response to pressures from interest groups. The broader issue of the distributional implications of supporting large-scale rather than small-scale farming, and the gender dimension, in the context of fragmented property rights that limit Moldova’s ability to take full advantage of its agro-climatic and geographical advantages, will be analyzed in a dedicated piece in the Poverty Assessment (forthcoming). 2.5. Conclusions and Recommendations 54. The GoM has made progress in aligning the budget with its agriculture sector strategy. Allocative efficiency has improved with the reduction of large-size recurrent subsidies and the increase of investment support and smart subsidies. The budget also reflects the priorities announced in the National Sector Strategy. 24 Similarly, 13 percent of all incorporated farms led by females have a bank loan versus 14.4 percent of male-led farms; 0.1 percent of unincorporated farms led by females have a bank loan and 0.2 percent of male-led farms). 37 55. Structural reform of some agricultural services is needed to make them more effective and heighten their value for money. Inefficiencies still negatively affect the outputs and outcomes of budget-financed activities. Budget allocations need to be revised in recognition of the new roles and functions of the reformed institutions. Renewed budget allocations need an embedded focus on building the long-term efficiency and sustainability of agricultural services through enhanced capital allocations. These findings suggest a need to continue refining the composition of the budget to change the quality and quantity of public services. 56. Given the inefficiencies affecting them, the government is invited to consider reducing some programs: for example, cancel or redesign the less effective credit subsidy program and the land consolidation program if uptake continues to be small. Further, reconsider the allocation to education if there are no reforms. Some other programs are essential but require revision to make them more efficient before increasing their budget share : (1) link research institutes with the private sector and the international R&D system; (2) revisit the irrigation strategy to create a model in which Water Agencies and Water User Associations share maintenance and rehabilitation of irrigation systems; (3) reform the Seeds Commission testing program to both better serve private sector needs and rationalize annual spending on these public goods; (4) increase the share for risk management tools preceded by improvements to the current system; (5) continue to reform the food safety programs. Finally, some programs warrant an increase because they are both critical and have so far achieved positive results: (1) build up the extension services, the budget share of which is small; (2) invest more in the post-harvest infrastructure program and consider further reducing subsidies for large machinery; (3) continue designing smart subsidies. 57. Finally, to facilitate and discipline the process of preparing the agricultural budget for institutions performing services in the sector, it is necessary to closely link it to annual performance evaluations of the institutions. This could be done by operationalizing inclusive M&E frameworks and setting clear rules and procedures for utilization of the M&E frameworks in the annual budget process. 38 Chapter 3. Tax Expenditures in Agriculture Agriculture receives favorable tax treatment. This reflects the large role of the sector, the high level of capital investment required—and the limited availability of financing—and the need for it to compete successfully in the EU market. Supporting rural incomes is also a government priority as it advances structural transformation. A large proportion of farmers are thought to fall outside the tax net because of preferential tax treatment arrangements, including a lower corporate income tax rate, lower social security and health contributions, concessions for citizens acquiring property, property tax concessions and a complex system of VAT concessions. All these tax expenditures amount to 0.6 percent of GDP, which is sizable in relation to explicit spending and tax collection in the sector. On the other hand current tax expenditures do not effectively address challenges identified for the sector. As suggested by good practice according to the theoretical and empirical literature and the experience of other countries, structural impediments in Moldova’s agriculture sector—and the country’s economic transformation—can be tackled in more efficient, more cost-effective, and more equitable ways. This requires a range of both tax and non-tax incentives, and removal of disincentives, related to capital investment, farm consolidation, land utilization, and income support. This chapter discusses how the tax system can be used to achieve the government’s objectives while minimizing the burden on government finance and the distortion to the efficiency and equity of the tax system. The chapter first reviews the favorable tax treatment of the sector. That is followed by estimates of the related fiscal cost of the main tax preferences. The next section discusses how effectively current tax expenditures address the challenges of Moldova’s agriculture sector; their economic and social impact is compared to good international practice from the perspectives of both research and the experience of other countries. Finally, the chapter suggests changes to the tax and social security systems to address structural impediments in Moldova’s agriculture in a more sound and fiscally responsible way. 3.1. Taxation of Agriculture in Moldova 58. Agriculture is one of the least-taxed sectors in the economy. Taxes collected from agriculture amounted to 0.8 percent of GDP in 2013, or 3.3 percent of agriculture’s gross value-added. As a result, agriculture competes with construction, housing services, and manufacturing for the least- taxed sector. Since 2008 the amounts collected from agriculture have plunged, mainly due to lower VAT revenues (Figure 3-1). This is the result of new lower VAT rates on several agricultural products. 39 59. The tax system currently gives Figure 3-1: Taxes Collected from Agriculture by Tax Type, farming favorable treatment. The 2008–14 farming sector benefits from several (Percent of GDP) favorable tax arrangements, which as an instrument of policy can be considered an alternative to direct spending. These so-called tax 25 expenditures result in reductions in tax revenue and constitute a fiscal advantage conferred on farmers individually or the agriculture sector as a whole. The main preferential tax treatment arrangements available to Moldovan farmers are: Source: MoF.  A reduced CIT rate: Farmers who operate through companies (joint stock or limited liability companies or agricultural cooperatives) are subject to a preferential 7 percent CIT rate; the standard rate is 12 percent. The current rate is up from the 0 rate that applied for 2008–11.  A reduced rate for social security and health contributions: The Social Security Fund and the Health Insurance Fund offer concessions to farmers: o Employers in agriculture (those who during the budget year engage exclusively in certain activities stipulated in Classification of Economic Activities of Moldova) enjoy a contribution rate of 22 percent instead of the standard 23 percent. Furthermore, 6 percent of an employee’s salary is transferred as a direct subsidy from the state budget, bringing the effective rate down to 16 percent. The social security system for farmers is managed with that for the rest of the population. Since contributions do not fully fund benefits, the government covers the gap. The government also funds from general revenues the main (non-agricultural) system, which is also in deficit. o Concessions for health insurance contributions are available for specified categories of taxpayers, including the unemployed who own agricultural land and individuals who contract to lease or use agricultural land. Thus, like individual farmers, these taxpayer farmers pay an annual flat premium for health insurance (MDL 4056 [EUR 197.92] for 2014–15) instead of 4 percent of gross salary. 25 In this study tax expenditures are defined as tailored reductions in tax liability (see Annex 7). 40  Concession for property acquisition by citizens. Transfer by gift or inheritance of real property, including agricultural land, is free of tax, but agricultural and forest land cannot be bought by foreign citizens, companies, or states, though they can lease it for the short term (1–3 years) or the long term (5–99 years). However, Moldova does not have special rules for capital gains tax in connection with agriculture. The gain, taxed at normal rates, is calculated as 50 percent of the difference between the sale price and the purchase price (all costs related to acquisition of capital assets).  A property tax concession. Concessions on taxation of agricultural property are complex but generally consist of low rates for agricultural land (Annex 6) and exemptions from the real property tax if the land meets certain conditions.26 Currently, Moldova is in transition from the old system of real property taxation based on inventory value (consisting of two taxes: a land tax plus a tax on immovable property, a real estate tax27) to a new ad valorem system. In the new system, both land and buildings are subject to a single tax, which is based on the market value of the property. The new system is not yet completed with only certain types of property covered; in practice, only the old system is in operation. Since 2013 the government has been working on establishing a new method to value immovable property for purposes of taxing it in the new system and expanding coverage of the valuation system to agricultural land and residential property in rural areas.  Complex VAT concessions, including a reduced rate. o A few Moldovan agricultural businesses have been given explicit concessions on their tax obligations. The main concessions have been a deduction for VAT input associated with the purchase of fixed assets for agricultural purposes, a VAT exemption for tractors used in agriculture, and VAT refunds for producers of agricultural goods subject to a lower VAT rate. Whoever pays the VAT is entitled to a refund if the company carries out one of the business activities specified (e.g., export supplies, international transportation services, production of bakery and dairy products, leasing activity, and supply of certain self-manufactured agricultural products, such as sugar). Before registering for VAT purposes, it is possible to deduct input VAT, but only if incurred on acquisition of fixed assets for agricultural purposes and construction and reconstruction of agricultural goods. This allows agricultural enterprises that invest in fixed assets to recover their investments faster.28 o A reduced VAT rate for agricultural products is an implicit concession to farmers. The standard VAT rate in Moldova is 20 percent, but a reduced rate of 8 percent applies to bread and 26 At least one of three conditions must be met: the land is (1) used for multiannual crops before they are harvested; (2) used for agricultural purposes for 5 years and was recognized as eroded when acquired and later was re-cultivated and restored; or (3) subject to chemical, radioactive, or other type of pollution if the government restricts agricultural activity on the land. 27 The land tax is calculated on the unit area of land and adjusted to reflect land location and fertility; the real property tax depends on the type of owner or use. Residential buildings are taxed on their inventory value (cost) and buildings used by businesses on their book value. 28 However, Moldova does not have a simplified VAT system like the one the EU makes available to some farm businesses. The rules for VAT registration are the same across the board: businesses can register for VAT purposes if total turnover within the last 12 consecutive months reached the threshold of MDL 600,000 (about €29,278.43). They also allow for optional registration as a VAT payer when the business supplies goods or services in an amount exceeding MDL 100,000 (€4,879.74) and meets certain additional criteria. 41 bakery products, milk and dairy products, and primary agricultural production and sugar (see Table 3.1.). Reduced VAT rates for agricultural commodities provide an indirect financial benefit to farmers, though consumers are likely to benefit more from this type of concession than farmers (a discussion of the economic impact of reduced VAT rates is presented below). Table 3.1: Value-Added Tax (VAT) Rates 20% Standard ratefil services, sale shares, sale or rental of dwellings and land, et 8% Certain types of supplies (e.g., bread, bakery products, milk, sugar, gas, certain pharmaceutical and 0%* phytotechnical Certain supplies products) (e.g., export of goods or services), international transportation, supplies to and from Exempt the territory Certain of Free supplies, Economic financial Zones, services, import sale or supply of shares, goods of of rental sale or and servicesand dwellings destined for technical land, etc., and assistance * VAT-exempt tractors with the projects used agriculture in to right under specific exercise conditions. the input VAT deduction. 60. Although in theory farmers are largely subject to the same income tax regime as other individuals, a large proportion are thought to fall outside the tax net . In Moldova, 58 percent of agricultural land is farmed by companies and 42 percent by small-scale enterprises. The latter (individual entrepreneurs, subsistence farmers, family farms) are subject to progressive income tax like other taxpayers. Farmers and individual entrepreneurs who are not registered as VAT payers may maintain a single-entry accounting system29 without a need to present financial statements, provided that the system is a method approved by the Ministry of Finance (MoF). Those entitled to use a single-entry system may still opt to use double-entry accounting. Farmers who met specific criteria30 in the previous reporting period are entitled to organize and maintain accounts using the double entry accounting system with a simplified presentation of financial statements. However, a large proportion of farmers are thought to fall outside the tax net. Only a fraction of the 550,000 individual farmers pay income tax, though statistics suggest this number is probably an understatement. The agriculture sector also constitutes a large part of an unobserved economy (about 60 percent of its VA or about 7 percent of GDP). 3.2. The Fiscal Cost of Agricultural Tax Expenditures 61. Tax expenditures in agriculture are sizable in relation to explicit spending and the taxes collected from the sector. Although defining and quantifying tax expenditures is not easy (see Annex 7), an estimate of tax expenditure points to foregone revenue of over 0.6 percent of GDP in 2014,31 which is sizable compared to both the taxes collected from agriculture and public spending on the sector. The ratio of tax expenditures in agriculture reached about 60 percent of the tax collected in agriculture and about 40 percent of direct budget spending on agriculture. 29 Single entry is often referred to as the cash method of accounting. In this relatively simple method of accounting, items are listed as income or expenses when cash transactions occur. For example, grain is recorded as income when it is converted to cash—that is, sold and delivered. Also, production inputs like seed, feed, and fertilizer are recorded as expenses when they are paid for rather than when they are used or ordered. 30 The three main criteria are (a) an average number of employees that does not exceed 9; (b) income from sales of less than 3 million lei; and (c) the total balance of assets carried over at the reporting date of less than 3 million lei. 31 The lack of statistical information, appropriate databases, and economic models for estimating tax expenditures in Moldova was a limitation to the exercise. 42 62. Tax expenditures are skewed toward special VAT treatment of agricultural output, which is a drain on the budget. VAT has the largest share in the tax expenditures granted to the sector— about 0.5 percent of GDP—while tax expenditures from direct taxes (together with SSCs) represent about 0.1 percent of GDP (Table 3.2). Corporate tax concessions are the second largest, followed by concessions in payment of social security contributions. Table 3.2: Estimated Cost of Tax Expenditures in Agriculture (in percent of GDP and percent) Taxes and Contributions Tax Expenditure, Share in Total Taxes from Percent of GDP Agriculture, Percent Reduced CIT rate 0.07 9.7 Reduced VAT rate 0.50 59.0 Reduced social security rate 0.05 6.1 Reduced health contribution rate 0.01 1.2 Total 0.63 63.0 Source: MoF data. 63. For VAT, the potential shortfall of revenues stemming from the reduced rate for agricultural products could be decomposed into a policy gap (the measure of tax expenditures) and a compliance gap. The policy gap for agriculture, which reflects the extent to which consumption of agricultural products is not taxed at the standard rate,32 is high at about 0.5 percent of GDP. The gap is driven by the reduced rate for VAT on consumption but also refunds of input VAT granted to the sector. The compliance gap reflects the effectiveness of tax policy administration in implementation of the VAT33 and consists of the amount of tax declared but not collected and the amount of tax due that was not declared or assessed. For Moldova the compliance gap is much lower than the policy gap at about 0.1 percent of GDP. Both gaps were calculated based on the RA- GAP VAT gap estimation methodology (see Annex 8Error! Reference source not found.). Unfortunately, the fiscal costs of other VAT concessions offered to farmers (deduction for VAT input associated with the purchase of fixed assets) have not been estimated due to data limitations. 64. The estimated size of agricultural tax concessions in income taxes and social security and health contributions is relatively small and mainly reflects reduced rates . Reduction of the tax rate for corporate farmers (legal persons) reduces state revenues by about 0.7 percent of GDP. In 2014 the state budget transferred MDL 44.3 million (0.04 percent of GDP) to the social security agency, CNAS, as compensation for a 6 percent lower contribution. The estimated implicit subsidy from the lower general rate (22 percent vs 23 percent) was MDL 7.3 million (0.01 percent of GDP). 32 The policy gap is zero if the VAT is applied at a single rate to all (and only) consumption. The calculation assumes full compliance. 33 The compliance gap is zero if implementation of the VAT is perfect. 43 Concessions in health contributions are small. In an extreme scenario in which all farmers that pay lump sum contributions switch to paying 8 percent of their gross salary (which is assumed to be twice as high as the average agricultural salary), the tax expenditure equals 0.1 percent of GDP. Unfortunately, data limitations do not allow calculation of the costs of current concessions in taxing land and other immovable property, capital gains, and the transfer of assets. 65. However, eliminating some, or even most, tax expenditures may not increase tax collections by a corresponding amount. At first glance, after reviewing the tax expenditure estimates, it appears that the government could collect more tax revenue by eliminating some, or even most, of those granted to the agriculture sector. However, such calculations need to take into account economic and behavioral effects. The costs of tax concessions also need to be set against the benefits they create (discussed below) and against missed opportunities to use public funds for other purposes or in a different way (e.g., through direct spending on the sector). 3.3. The Economic and Social Impact of Tax Expenditure in Agriculture 66. As an instrument of policy, tax expenditures should help the government to deliver on its objectives for the sector. Tax expenditures granted to agriculture can be viewed as having one or more of the following aims: (1) support the competitiveness of the agricultural sector through restructuring and modernization (e.g., provide economic incentives to support investment in fixed assets or more productive use of land); (2) incentivize sustainable management of natural resources; and (3) help improve conditions for those living and working in agricultural and rural areas (e.g., provide financial assistance to certain groups of farmers and their families, without limiting structural change in the sector). An effective tax system—one that raises the revenue needed for core governmental functions while minimizing the burden on citizens—should also follow some basic principles (Error! Reference source not found.). Do tax expenditures live up to the expectations of them? The following section assesses potential benefits and distortions created by the tax expenditure system in Moldova.34 67. The favorable tax treatment of income is likely to slow restructuring of the agricultural sector. Reduced rates for health contributions for individual farmers and reduced CIT rates and SSCs for corporate farmers are likely to bear on decisions of farmers and landowners to stay in the industry and slow the much-needed structural transformation. The income taxation rules for farmers do not distort the choice between being organized as a company or as an individual farmer for small-scale farmers (annual taxable income of up to MDL 27852) since up to that level of income they offer the same personal and corporate tax rates. In the short run these preferences support corporate profits and employment in agriculture. In the longer run, however, lower effective taxation of agricultural incomes can stimulate reallocation of resources toward agriculture and affect decisions in favor of diversification. Lower SSCs for farmers will have a similar effect, and lower employer 34 The economic and social impact of tax expenditures in Moldova is assessed by comparing them to (1) good practice in taxing agriculture derived from a review of the theoretical and empirical literature (mainly OECD 2005) and (2) the experience of other countries with selected tax expenditures (both advanced countries like Ireland, the U.K., the U.S., and Germany and also Moldova’s comparators, such as Bulgaria, Romania, and Ukraine). 44 contributions for hired workers will keep more of them in the industry. Preferable treatment of agricultural income in any case is not equitable and generates high costs. Finally, the income tax concessions do not provide support to low-income rural households or reduce the volatility of farmers’ incomes (e.g., through farm income averaging). 68. Moreover, in contrast to practices in other countries, Moldova’s taxation of incomes favors corporate over individual farmers. Compared to both advanced countries and countries in the region, Moldova’s PIT concessions do not look very generous. Several countries in the EU use a special system of assessment in taxing individual farmers (see Annex 9), and Georgia, Bulgaria, and Romania exempt from taxation in whole or in large part the incomes of individual farmers (see Table 3.3). Presumptive taxation is also practiced (e.g., in Belarus). In the EU farmers are also the main group using the averaging provisions in income tax systems or income equalization funds. All of these PIT concessions attempt to deal with income instability, low incomes, and the associated poverty of individual farmers and their families. Special treatment of farming companies is less frequent, and here Moldova’s reduced CIT rate stands out as an exception. Several advanced countries have special arrangements for SSCs in agriculture, but deciding whether these constitute concessions is difficult because the social security system for farmers is typically run separately from that for the rest of the population. 69. Moldova’s current system for taxing real property may both delay exit from agriculture and block entry into it. The area-based system of land taxation in Moldova is comparatively simple to administer and manage. However, an area-based property tax system is often perceived to be unfair because area-based rates bear little relationship to either ability to pay or benefits received, e.g., there is very little difference between the tax rates for land in the capital and in other cities in Moldova. Agricultural real estate is exempted real property tax if, for instance, the land is used for multi-annual crops before their harvest. This exemption from taxation, together with the tax concession on transfers of agricultural property, may delay exit from agriculture and raise impediments to entry, particularly for foreigners. The evidence from Romania and Bulgaria shows that restrictions on foreign ownership have affected the efficiency of land exchanges, land allocation, and productivity growth there. However, as in Moldova, agriculture receives concessions on annual real estate taxes in most advanced and developing countries. 45 Table 3.3: What Is Taxed Where, Moldova and Regional Comparators Income (personal) Profits (companies) SSCs Health Property Care Contrib. System* Rates Other Rates Other Employer Employee Capital Inheritance Annual taxes (specify) (specify) gains on assets Bulgaria No No Yes, Income No No No Yes Yes from agriculture non-taxable Georgia No No Exempt if No Selected Does not Does not No na Yes. income from types of have SSCs have SSCs agriculture profit (e.g., <200,000 primary laris production) Moldova No No No Yes, No Yes, No, but Yes, flat No Yes Yes reduced reduced paid by the payment rates state Romania No No Yes, in large No No No No No No No Yes not taxed Ukraine Yes, for No No No Simplified No No No No No Simplified agricultural regime; tax regime; producers with base is area replaces land sales < UAH of tax as well 20M (USD 800 agricultural k) land Source: Interviews with respective country experts. Note: * Special system applied to agriculture. 46 70. VAT concessions on agricultural inputs are not justified by market failures and are likely to have an economic cost because they may induce a change in the pattern of production. The impact of the tax concession on agricultural inputs (e.g., investment and tractors) is similar to that of subsidies on agricultural inputs. These Figure 3-2: Investment in Long-Term Tangible concessions alter the prices of inputs relative to what other users pay, leading to inefficient Agricultural Assets allocation of inputs and a change in the (Percent of GVA) pattern of outputs.35 Although use of input Investments in long term tangible assets 10 Private 0.4 support is often motivated by the need to Public RHS Foreign investors RHS address market failures and social objectives, 8 0.3 it is difficult to argue that is the case in 6 Moldova. Finally, concession on inputs are 0.2 4 prone to abuse because they encourage 0.1 farmers to divert exempt purchases to VAT 2 exemptions and thus may benefit unintended 0 0 recipients. The actual impact of such a 2006 2007 2008 2009 2010 2011 2012 2013 measure is not clear—data on investment in Source: Stat data. agriculture shows that private investment had gone up in 2011–12 before the VAT concessions were granted (Figure 3-2). 71. Although a reduced VAT rate for agricultural products is quite common in European countries, the efficiency and equity rationale for supporting that approach does not apply in Moldova. Extending reduced VAT rates to food affects the efficiency of a VAT system and increases complexity, which in turn increases administrative costs (for system operation and monitoring) and compliance. Compliance costs seem to be particular high for the food sector due to its multitude of products and the gray zone between sales of foods themselves and of prepared food. However, there are economic arguments for granting lower rates to very specific sectors, such as agriculture, and their theoretical and empirical merits need to be discussed. These arguments are based on grounds of both efficiency (heightening productivity or reducing structural unemployment) and equity (improving the income distribution or making particular products more accessible to the entire population (see Annex 11: . In practice, a reduced VAT rate for agricultural products is quite common in European countries. However, some of these arguments do not hold for Moldova; for example, the reduced VAT rate on certain agricultural products is in general regressive because it accrues mostly to higher-income quintiles (Figure 3-3).36 Thus, it is questionable whether the reduced VAT rate is the best instrument for improving income distribution, especially in countries like Moldova that have sound social security systems. 35 An impact similar to that of the VAT concession would come from lowering fuel taxes for agricultural purposes—something now being debated in Moldova. All else being equal, such a reduction could increase fuel usage and the production of agricultural goods and services, not only market output but also non-market environmental goods and joint products. 36 While the share of consumption spent on food and agricultural items whose VAT is subsidized is relatively flat (from 12.2 for the bottom quintile to 13.4 for the 4th), richer groups consumer vastly more of these goods than poorer ones (over 33 percent of total spending on these goods comes from the top quintile, against 10 percent from the bottom quintile). 47 Figure 3-3. Benefit of the VAT Discount, by Quintile, 2013 (Percent) 0.40 33.9% 0.35 0.30 24.2% 0.25 0.20 17.9% 14.1% 0.15 10.0% 0.10 0.05 0.00 1 2 3 4 5 Source: WB staff calculations 3.4. What Might be Changed? 72. The taxation and social security systems could be used to tackle agricultural problems in ways that are more efficient, more cost-effective, or more equitable. More effectively addressing structural impediments in Moldova’s agriculture sector requires both tax and non-tax incentives— or removal of disincentives—related to capital investment, farm consolidation, land utilization, and income support. Incentives to Invest in Agriculture 73. A proper investment structure is vital to support private investment and should be in place before considering specific tax incentives. The experience of other countries suggests that tax incentives for investment can be costly and are rarely the main determinant of whether to invest. That is why Moldova needs to focus directly on improving the environment for enabling investment to ensure that there is a proper context for investing in agriculture before considering specific incentives.37 Tax incentives to support investment include reductions in the statutory CIT rate; accelerated depreciation allowances for capital expenditures; enhanced depreciation allowances for capital expenditures; general or targeted investment tax credits; and financing incentives (see Annex 13). Capital allowances (accelerated depreciation or an investment tax credits) are likely to be more efficient and have fewer drawbacks than other types of tax concession (e.g., tax 37 The OECD Policy Framework for Investment in Agriculture (PFIA), which aims to support countries evaluating and designing policies to mobilize private investment (OECD 2013) could be a useful guide with several aspects relevant for Moldova, including infrastructure development, financial sector development, human resources, research and innovation, tax policy, and risk management. 48 exemptions), and in many countries they are replacing income tax exemptions or reduced CIT rates.38 74. Tax expenditures in Moldova could be redesigned to make them more effective in supporting investment and to increase the value for money in the use of public resources. The authorities could consider increasing the CIT rate for agriculture from 7 to the standard 12 percent, which would still be one of the lowest CIT rates in the world. The reduced CIT rate gives the largest benefits to high-return firms that are likely to have invested even without incentive. Moldova’s recent experience with a zero CIT rate has also demonstrated that the low rate did not have any incremental impact on investment in the country. Moreover, given its distortive nature and the lack of visible results, the authorities could consider removing the VAT exemption for machinery and tractors. The revenue gains from these changes could be used to offer more effective instruments to support investments in agriculture, such as a capital allowance or tax credit similar to those used in more advanced countries. For instance, capital allowances for expenditures could be directed to improving the productivity of land or favoring environmental (soil-friendly) investments.39 Thus, with a CIT rate of 12 percent, an investment allowance of 50 percent of the amount invested equates to an investment credit of 6 percent of that amount and revenue foregone of about MLD 100 million. Incentives for Better Utilization of Land 75. Better utilization of land, a crucial asset for agriculture, could make a significant difference in how the sector performs. There are several ways in which government might incentivize better land use, in particular by allowing foreigners to acquire land, adopting value-based taxation of property (which requires effective valuation of land and property), and taxing uncultivated land. 76. Abolishing the limitation on land acquisition by foreigners can open the sector to new entrants who might bring in new resources and technology. Here, the experiences of Bulgaria and Romania are particularly useful (Vranken 2010), One option could be to allow foreign citizens to acquire agricultural land without restrictions up to a certain area (less than 10 ha) and allow foreign citizens to acquire farm buildings and the land on which these are built without restrictions. Both proposals could result in substantial economic benefits as foreign citizens investing in the agricultural sector in Moldova could combine renting and owning land in their farm operations, as do many farms in the EU. 77. It would also be advisable to complete the process of revaluing agricultural land and property and to introduce a transparent value-based system for taxing property. Here, Lithuania provides a good example of a well-developed system of mass valuations with periodic revaluations using 38 For instance, the evidence from Ireland is of an indicative impact on gross output of €1.9m for every €1m of capital allowances claimed. 39 Of course there are several design issues of this kind of tax incentives that need to be carefully thought-through to make it work. Main include the definition of the eligible expenditures, the choice of the rate of the allowance or credit restrictions on the use of the credit or allowance, the payable element of the credit and the treatment of any amounts of incentive that cannot be used in the year that they are earned as a result of insufficient taxable income. 49 different models for different types of property. Currently, the insignificant income from land taxes in Moldova is viewed as a factor preventing reallocation of agricultural lands to more efficient users, which leaves large areas uncultivated. 78. Also worth considering might be introduction of a special tax on uncultivated land to incentivize better use of it, as has been done in Latvia, Croatia, and Romania.40 Introduction of a tax on uncultivated land in Moldova of about MLD 100/ ha could bring some in MLD 45 million for the state budget annually (assuming that the amount of uncultivated land in Moldova totals 0.2 million ha and full compliance will be achieved). Support for Farmer Incomes 79. Given the impact of the agriculture sector on the poverty map in Moldova, it is essential to raise farmers’ incomes and ensure that they are stable. In the short to medium term, there is a need to raise the basic incomes of the poorest farmers. Most advanced countries use social policies to protect incomes and provide support for farmers who are unable to adjust with social aid to families concentrated in means-tested schemes requiring an application. In reality, however, those with low incomes from agriculture often fall below the tax net, and not being part of the general tax system they cannot receive social system benefits. That is one reason why incentives are needed to report income, along with reducing the burden of taxation for small farmers. 80. A presumptive tax regime would carry a lower tax burden and lower compliance costs. The incentive to report agriculture income in Moldova could be increased by introducing a simplified method of taxing individual farmers, with one option being a presumptive tax regime to replace PIT for this group. There are several types of presumptive taxes, such as. a patent, an indicator-based tax, a gross-basis turnover tax, and a net (adjusted) turnover tax, with a number of variants for each. The most common form is a turnover tax levied on gross revenues, which unlike a patent or indicator-based tax varies directly with the farm size. A turnover tax, besides addition to imposing a higher tax burden on less efficient farms, would tend to discourage allocation of capital to business activities where profit margins are relatively thin, thus not compromising equity and economic efficiency too much. Austria provides an example of this type of tax.41 In Moldova, the presumptive tax could be applied to agricultural producers with turnover below the VAT registration threshold, with cost deductions like those of Austria. Of course, the design of the tax should be non-distortive and help maintain equity with taxpayers covered by standard tax regimes. A presumptive tax is 40 For instance, Latvia applies a tax rate of 3 percent to uncultivated land capable of agricultural use, compared to the normal rate of 1.5 percent. The Romanian authorities intend to introduce a tax on uncultivated land that will oblige owners to work, rent, or lease their land, or sell it, though the proposed annual tax of about €100 ($138.30) for each uncultivated hectare h as yet to be passed. Croatia also applied a tax on uncultivated agricultural land between 2001 and 2007. 41 The tax is applied to unincorporated businesses only; the tax base is measured as turnover minus wages, cost of goods (inputs), and related taxes (including VAT on inputs), less “deductible expenses” measured simply as 12 per cent of turnover (up to a maximum of €26,400). The simplified regime applies to businesses under a turnover threshold that varies by business sector. 50 introduced in the simplest form could bring in about MLD 50 million (5 percent tax rate applied to gross profits). 81. Building up prospects for higher incomes should be put on the long-term reform agenda. Ultimately, Moldova will need to move from improving incomes from current activities to improving labor productivity and facilitating adjustment to activities that might command higher returns. This means not only skills upgrading, technology improvement, and improving off-farm employment prospects in rural areas but also limiting disincentives to move out of agriculture. As discussed earlier, social security and property tax concessions discourage farmers from leaving the sector. Thus, it is worth considering removing the SSC reduction for farmers, which would not only reduce disincentives for labor to leave agriculture but also make the pension system more sustainable. 82. Reintroducing a unified VAT rate for agriculture could help limit production inefficiency and administrative costs. The reduction of the rate was intended to make agriculture more competitive by reducing costs, with the objective of boosting activity in the sector and creating jobs. It does not seem to have been very successful. As evidenced by the Mirrlees Review (2011) it may also lead to both consumption and production inefficiency (the distortion arising from favoring the industry that has a lower VAT rate). Evidence from other EU countries is that administrative costs are high due to constant delineation problems linked to making distinctions between different kinds of food. Thus, Moldova should consider returning to taxing all food at a standard rate. Since the change would have negative impact on households with the lowest incomes (see Annex 12), part of the revenue gains should be used to compensate those most affected. Three groups that could be given priority for a compensation program are (1) beneficiaries of Ajutor Social, the current guaranteed minimum income program, which could add a top-up to compensate for the VAT increase; (2) the poor, since protecting them from further economic hardship should be a priority; and (3) the poor and other who have become vulnerable to poverty due to the VAT increase. With such focused targeting, the compensation would absorb a mere 6 percent of the budget currently absorbed by the reduced VAT rate. Thus, an increase in the reduced VAT rate for agriculture could not only limit production inefficiency and administrative costs but allow for cost-efficient income distribution. 83. The proposed changes to tax expenditures could bring the budget extra revenues. Several measures could bring in higher revenues, with the largest being to do away with the reduced VAT rate (see Table 3.4). Significant revenue gains would also come from a new land utilization tax if the government decided to levy MLD 100 / ha of unutilized land. The assumption of a rate of 5 percent on gross profit for a presumptive tax would give revenues of about 0.04 percent of GDP, though clearly the cost of the simplified system will depend on the tax rate and decision of who is in the tax base. At the same time, spending will increase by MLD 170 million, 0.2 percent of GDP, to cover introduction of programs to compensate those affected by the VAT change, a new investment tax credit/allowance, and the simplified method for taxing the incomes of individual farmers. The net effect of the proposed changes is likely to be additional revenues of about 0.5 percent of GDP. 51 Table 3.4: Budgetary Effect, Proposed Changes to Tax Concessions in Agriculture MLD millions Percent of GDP 1 Unified VAT rate for agriculture Increased VAT rate (8%-> 22%) 502.6 0.48 Spending on compensation -48.8 -0.05 2 Changes to investment incentives Increased CIT rate (7%->12%) 74.1 0.07 New tax credit -100.0 -0.09 3 Income support Introduction of presumptive tax 46.9 0.04 PIT revenue loss -65.7 -0.06 4 Land utilization Tax on unutilized land 44.9 0.05 5 Other changes Increased rates for SSCs 51.6 0.05 Total (net) impact 505.6 0.49 Conclusions and Recommendations 84. Tax expenditures for agriculture in Moldova have a serious cost and do not effectively address the challenges identified for the sector. They do not seem to have been effective in (1) providing income support for farmers to reduce their poverty; (2) fostering structural change in the sector and the whole economy; or (3) offering an incentive for the sector to become more competitive. Concessions granted on taxation of income property and social security encourage workers to stay in the sector and do not help attract new, more productive entrants. VAT concessions are not effective in attracting investment in fixed assets and equipment upgrades—if there are any economic efficiency gains of a reduced VAT rate, they are not evident. Instead, concessions involve substantial administrative costs. Moreover, the current tax system does not provide incentives for consolidation or better utilization of land, fails to heighten the market orientation of semi- subsistence farms, and does not address poverty in the sector. Most of the concessions violate at least one of the basic principles of a fair and effective taxation system; the special VAT arrangements for agriculture seem not to respect any of them. Finally, tax expenditures in agriculture represent a significant fiscal cost of about 0.6 percent of GDP, which constitutes about 40 percent of Moldova’s direct budget spending on agriculture. 85. The taxation and social security systems could be used to tackle agricultural problems in ways that are more efficient, more cost-effective, or more equitable with the net effect of the proposed changes likely to result in extra revenues of about 0.5 percent of GDP. Addressing structural impediments in Moldova’s agriculture sector more effectively requires a range of both tax and non-tax incentives—or removal of disincentives—related to capital investment, farm consolidation, land utilization, and income support: 52  Increasing the CIT rate for agricultural enterprises to the standard 12 percent and removing the VAT exemption for machinery and tractors would bring revenue gains from offering more effective tax instruments to support investments in agriculture, such as a capital allowance or tax credit similar to those used in more advanced countries.  Better use of land can be incentivized by (1) abolishing the limitation on land acquisition by foreigners, in order to the sector to new entrants who might bring in new resources and technology; (2) effective valuation of land and property, which is vital to adopting value- based taxation of property; and (3) introducing a special tax on uncultivated land.  At the same time it is vital to provide income support to the poorest farmers. Introducing a presumptive turnover-based tax, with a low tax burden and low compliance costs, would bring low-income farmers back into the tax net and make them eligible for means-tested social assistance. To strengthen prospects for higher incomes through sector restructuring it is crucial to limit the disincentives to moving out of agriculture and consider removing the reduction in the SSC rate for farmers. Last but not least, re-introducing a unified VAT rate for agricultural products (while ensuring proper compensation for low-income households) could help limit production inefficiency and administrative costs. 53 Annexes Annex 1: Modernization and restructuring of agri-food industry and adapting it to EU standards – the experience of Poland. Reaping the benefits of the trade integration with the EU depends to large extent on the ability to meet the body of Community legislation with the adoption of the acquis communautaire being the main priority of the EU accession process. For agriculture this includes the sensitive areas of veterinary and phytosanitary legislation and in eight CEE countries joining the EU in 2004 required significant investment in restructuring and upgrading the agri-food industry to make it competitive, and to comply with EU quality and sanitary standards. Especially, huge investments were necessary to adapt to EU standards in the meat and dairy sectors. Among the eight CEE countries joining the EU in 2004 Poland appeared to have most pronounced problems with adjusting to EU regulation, in particular regarding health rules of the production and processing of dairy products, where (as of 2003) national legislation was only partly in compliance with EU regulations (similar to Bulgaria and Romania). But also in other countries full implementation and enforcement of the acquis with respect to health rules of the production and processing of dairy and meat products faced some challenges due to a shortage of well-trained and qualified staff, an ineffective monitoring and/or penalty systems or unclear divisions of authority and competencies. Furthermore, the huge investments necessary to adapt to EU standards caused severe problems. In 2002 only 15% of enterprises in the dairy sector in Poland had a license to export to the EU with 35% of production satisfying the criteria for export to the EU – among the lowest shares in the EU8. The cost of complying with EU standards was estimated at 600m EUR and regardless of significant support being provided under the SAPARD and national programs considerable part was to be taken by the enterprises (with 25% of dairy enterprises and ca. 10% of capacity expected forced to close down because not being able to comply to EU standards). As for the meat industry the cost was estimated at 1bn EUR with ca. 40% of meat processing enterprises and ca 15% of capacity estimated to be forced to close down due to not being able to comply with EU standards. 54 Annex 2: Determinants of TFP in agriculture There is an extensive literature about the determinants of TFP in agriculture. Some authors (e.g. Bravo- Ortega and Lederman, 2012 or Mundlak, Yair, 2001) consider such measures as an illiteracy rate, paved roads or telephone lines as important determinants of agriculture productivity. Given significant correlation of such factors as the number of telephone lines or illiteracy rate with GDP per capita, in our analysis we focused only on agriculture-specific determinants of productivity. According to Cimpoies et al.(2008), three main problems of Moldovan agriculture are orientation to export, climate vulnerability and land fragmentation. The impact of land fragmentation on productivity was quite extensively researched and described it the literature – in general there is a positive correlation between farm size and productivity (e.g. Adamopoulos, Restuccia, 2011).However, for Moldova the relationship is different and small farms are more productive (see Lerman, Sutton (2008)) The reason for this difference is probably due to the fact that in Moldova large farms are, in general, big conglomerates, that were previously state owned farms and are equipped with obsolete materials and technologies and resist modernization. To see the relationship between country specific TFP rate (in our analysis we decided to exclude time variability as this is to large extend due to weather and country fixed effect as a proxy of country specific agriculture productivity. As the overall level of development (captured by the level of GDP per capita) was already included in the production function, we decided to exclude variables which correlate with them from the regression analysis and focused on the variables, which, theoretically should not be correlated with GDP. Therefore, we considered following variables:  Share of livestock in agriculture product (as a proxy for agriculture specialization);  Share of irrigated land;  Share of permanent crops area in overall agricultural area (land utilization);  The variability of agriculture output;  Share of export in agricultural production;  Average size of agriculture holding (a proxy for land fragmentation);  Orthogonalized government outlays on agriculture Some variables were included in logs if the distribution of values in the sample or theoretical reasons suggested to do so. The results of that exercise are presented in the table below. In general, they are not surprising, robust and immune to the changes in specification. Results clearly show that there are a few equally important determinants of agricultural productivity. The most important components are land utilization, irrigation, variability of agricultural output and government outlays on agriculture. Therefore, government outlays are only one way to increase the agricultural output, the other ways are increase in the utilization of land and reduce variability of output (possibly through irrigation). As almost all variables enter the equation in log (standard specifications were also tested but caused estimation problems) their impact is multiplicative, so increase in utilization 55 of land or drop in availability will increase the efficiency of agriculture spending. The result for the average size of agriculture holding is somewhat puzzling, as they are no clear correlation between TFP in agriculture sector and average size of agriculture holding. However, in almost all cases, the coefficients are positive. Also, share of export helps to explain the differences of TFP between countries, however, in that case, the causality runs the other way round – countries with higher TFP are more competitive on international markets so they manage to send more output to other states. 56 Table A2 Determinants of the fixed effect component from production function regression (Dependent variable: ui) 1 2 3 4 5 6 7 8 9 10 11 12 Share of livestock in agriculture 1.469*** 1.553*** 1.375** 0.724* 1.116*** 1.855*** 1.241** - - - - - production (0.502) (0.482) (0.564) (0.411) (0.404) (0.52) (0.541) log(share of irrigated land in 0.184*** 0.179*** 0.15** 0.146*** 0.127** 0.12** 0.203*** 0.181*** 0.198*** - - - agriculture area) (0.049) (0.049) (0.057) (0.042) (0.048) (0.047) (0.045) (0.046) (0.049) log(share of land cultivated 0.148** 0.14** 0.215*** 0.143*** 0.255*** 0.194*** with long-term crops in - - - - - - (0.056) (0.055) (0.061) (0.05) (0.053) (0.058) agriculture area) Variability of agriculture -0.676*** -0.604*** -0.924*** -0.568*** -0.455** -0.809*** -0.693** -0.762*** -1.137*** output (coefficient of variation - - - (0.23) (0.201) (0.241) (0.193) (0.2) (0.243) (0.268) (0.279) (0.256) of agriculture output) log(Agriculture export as a 0.192*** 0.179** 0.209*** 0.241*** 0.192*** - - - - - - - percentage of output) (0.07) (0.066) (0.071) (0.06) (0.064) Orthogonalized 0.198*** 0.226*** 0.181*** 0.193*** 0.253*** 0.253*** 0.3*** 0.342*** 0.393*** 0.346*** log(government outlays on - - (0.054) (0.032) (0.043) (0.058) (0.062) (0.065) (0.067) (0.061) (0.068) (0.051) agriculture) log(average size of agriculture 0.044 0.062 -0.045 0.083 0.099* 0.144** 0.223*** 0.104* 0.211*** - - - holding) (0.051) (0.042) (0.048) (0.053) (0.058) (0.058) (0.053) (0.055) (0.055) Const 0.313 1.834*** 0.733* 1.683*** -0.057 -0.592 -0.176 -0.959** -0.926** -2.068*** -1.527*** - (0.479) (0.329) (0.372) (0.258) (0.496) (0.522) (0.514) (0.457) (0.442) (0.408) (0.255) R^2 76% 80% 66% 69% 53% 72% 64% 59% 52% 50% 34% 31% Adj R^2 72% 76% 62% 65% 50% 67% 60% 56% 49% 47% 32% 31% F-stat 18.2 22.9 15.1 20.3 15.0 17.2 14.9 15.7 16.1 21.3 17.2 45.2 N 48 48 53 62 72 48 48 48 48 69 69 101 ***-statistically significant at 1% level, **-statistically significant at 5% level, *- statistically significant at 10% level 57 Annex 3: Local mapping of Agriculture Expenditures The local mapping for the agricultural sector shows that the institutional and financing structure for agriculture is complex.42 The Ministry of Agriculture and Food Industry (MAFI) is the principal government body responsible for funding and administering agricultural policy. Between 2009-13, the budget of MAFI and its subordinated agencies/deconcentrated regional offices accounts for approximately 88 percent of the sector budget on average. A significant number of subordinate agencies and institutions 43 are not directly part of MAFI but under its oversight and funding, which complicates effective management of the funds allocated to agriculture. 44 Public services in agriculture that are non-MAFI managed include mainly agricultural research activities as part of the Academy of Sciences budget for vocational school (6.6) percent and irrigation (2.4 percent) administered by the Ministry of Environment for irrigation. In addition to these, there are some agriculture-related expenditures under the Cadaster Agency for soil fertility activities, approximately 2 percent. However, it is an extremely small amount and it is not included in the estimates due to difficulties in separating the recurrent and capital agriculture-specific allocations. Lastly, the share of local spending in the sector budget is negligible (less than 1 percent over 2009–13). 42 The local mapping is different from the COFOG definition but it captures the amount allocated to the agricultural sector only, excluding forestry, fisheries and other non-directly related spending. The local mapping will used for any internal and sector comparison. 43 The total number of active institutions amounts to more than 25 plus 20 VET schools 44 MAFI is represented by its core central unit along with other agencies/inspectorates, such as “Main State Inspectorate for Phytosanitary Quarantine”, “State Inspectorate for Seeds of RM”, “State Inspectorate for Selection and Reproduction in Animal Husbandry”, “Consolidated Unit to implement the International Fund for Agricultural Development”, etc. At the sub national level, MAFI is represented through its deconcentrated offices (Food Security Directorates). 58 Annex 4: Agricultural Support 2014 Amounts paid (US$) & % # Measure Type Targeted Subsidy details share of total of sectors fund, 2014 subs M1 Credits (both Rec Cross-sector Subsidy amount approx. equivalent 1.13 / short- and to interest rate amount (calculated medium-term) based on a formula) and cannot be 2.9% higher than the interest rate. -Max subsidy MDL200,000 per credit for bio agriculture; -Max MDL100,000 per credit for other purposes. M2 Risk Insurance Rec Cross-sector -50% of insurance premiums for all 2.09 / crops and animals; The beneficiary pays 50% of the 5.4% insurance premium to the Insurance Company, which claims the remaining amount directly from AIPA. M3 New orchards & Inv HVA: fruit, nuts, -Subsidy amount varies between 6.69 / vineyards & berry berries, grapes, MDL10,000 and 50,000 per hectare + anti-hail & anti- aromatic plants depending on crop type and tree 17.3% frost systems density. -MDL50,000 for anti-frost systems; -MDL80,000 for anti-hail systems; Young farmers +10%. M4 Protected field Inv HVA: vegetables -60% for individual agricultural 3.63 / vegetable grown in producers; production protected field -75% for producer groups. 9.4% [greenhouses, -Max subsidy per beneficiary tunnels, solar.] MDL5m. Young farmers, organic producers +10%. M5 Machinery & Inv Cross-sector -50% for irrigation equipment; 8.42 / Equipment, -25% (& max MDL500,000 per unit) including for various ag machinery and 21.7% irrigation equip. equipment; 59 -Max MDL750,000 per beneficiary. M6 Livestock Inv Livestock sector -60% for cattle farms; 50% for 3.36 / equipment sheep/goat farms; -30% for other livestock farms; 8.7% Max MDL5m per beneficiary. Young farmers +10% M7 Livestock breeds Inv Livestock sector -MDL20-100 per unit [kg] of animal 3.38 / weight (amount depends on the type of animal) 8.7% -The min number of animals eligible for subsidy: 3 cows, 5 sows, 10 sheep… M8 Post-harvest and Inv Cross-sector, -50% for cold storages and packing 10.07 / processing mainly HVA houses; max MDL4.5m per equipment beneficiary; 26.0% -40% equipment for processing, drying and freezing of plant products ; max MDL2.5m per beneficiary; -30% equipment for processing of meat and dairy products; max MDL 1.5m per beneficiary; Young farmers, producer groups +10%. M9 Land Rec Cross-sector -50% of fees associated with merging 0.003 / Consolidation of at least 3 land parcels; 0.0% Max subsidy per beneficiary MDL100,000. M10 Energy for Rec Cross-sector, Open to individual farmers & WUAs: 0/ Irrigation mainly HVA -50% of energy costs for pumping 0% water from Central Irrigation Systems nd (CIS) & 80% for 2 time pumping; -MDL0.5 per m3 for pumping water from other sources; 60 Annex 5: Agricultural Education and Research Institutions Public Services in Agriculture The agriculture education budget is allocated approximately 50/50 between the 8 agricultural colleges and the Agricultural University. There are also about 20 vocational schools that teach agricultural subjects, but these are funded through the Ministry of Education budget (not part of MAFI budget). The Agricultural University offers courses in about 25 specialties ranging from traditional agricultural technical subjects such as agronomy, horticulture, animal husbandry to more general subjects such as economics and law. Agricultural colleges offer more narrow specializations, such as accounting, food technology, viticulture, mechanization. However, the curricula for most of the courses taught in agricultural education institutions had undergone little adjustments since old times. Even though education expenditures are the largest PEA budget item, these are considered highly insufficient to ensure adequate development of the agricultural education system, including teachers’ trainings, updating of curricula, repair of buildings, upgrading of equipment and other needs. The image of VET schools in the Moldovan society is particularly negative and is reflected in the decreasing number of students. Considering that most of the agricultural population is presently represented by family farmers the level of specialization taught in VET institutions is way too narrow. The young people trained in agricultural VET schools presently do not receive the kind of training that would help them become successful polyvalent farmers as small-scale entrepreneurs. More than 20 years after the dismantling of the collective farm system in agriculture and appearance of small family farms in Moldova, there is still no curriculum developed for training of self-employed farmers entrepreneurs. Ag research expenditures are part of the Academy of Sciences budget and come to finance the research activities of six institutes and the Agricultural University. Each research institute is in charge of a certain agricultural sub-sector/ area (such as field crops, horticulture, soil protection, animal breeding), while the research area of the Agricultural University is somewhat broader. Apart from research activities, some institutes are heavily involved in commercial activities (seed and seedling production, multiplication and sale) these activities bringing additional extra-budgetary revenue to the institutes. The agricultural research system has not managed to break with the past and reconnect with the private sector, it still operates in relative isolation and is fairly weak. Under- funded, struggling with dilapidated inventories and inadequate resources, aging of staff (due to low salaries), the existing research institutes are in a survival mode. There is no clear procedure/process established on consulting the selected research topics with the end-users of the research, i.e. representatives of the farming and business community. The risk of irrelevance appears particularly critical as long as the research institutes continue to operate in isolation from the private sector and from the international R&D system. The interlinking of research and commercial activities is not a healthy situation either, leading to inefficiencies and distorted public policies (e.g., research institutes lobbying for restrictive access to imported plant varieties thus undermining farmers’ interests). The core food safety budget is allocated to the National Food Safety Agency (ANSA), the single food 61 safety authority in charge of ensuring the control over the entire food chain “from farm to fork”. ANSA was set up in 2013 and was the result of merger of several public institutions previously responsible for various aspects of food safety and quality management. This institutional restructuring does not seem to have led to budget savings so far (the 2013 ANSA budget was similar to pre-2013 annual budgets allocated to food safety institutions). ANSA presently employs about 1500 people and disposes of a central office, 35 regional offices and 18 border control points. ANSA, strongly supported by donors, is presently pro-actively engaged in harmonizing its legal and institutional framework with EU practices and requirements, and significant progress has been made in various areas of food safety and quality control. However, significant efforts are still needed to set up a robust, reliable and effective domestic food safety system comprising the key elements of: (i) bringing food safety requirements and norms in line with internationally accepted practices; (ii) optimizing the laboratory setup, improving the reliability of lab testing and achieving laboratory accreditation; (iii) transitioning to risk-based inspections and controls; and (iv) setting up advanced information and data management systems. The Anti-Hail Service is in charge of organizing and executing active cloud influence works aimed at reducing the damages caused by hail. This is a large structure employing over 1400 people comprising of a central office, 12 regional units, and within each regional unit there are several ‘rocket points’ equipped with anti-hail launchers, rockets, storages and dwellings. The anti-hail coverage is extended mainly to high-value production areas (such as orchards and vineyards), while annual anti-hail allocations may vary significantly depending on hail propensity forecasts/conditions on the covered areas. Earlier World Bank studies45 indicated that there is no formal research demonstrating the cost effectiveness of such a hail prevention system, and that a cost-benefit analysis should be carried out to compare the efficiency of the current anti-hail rocket approach to a risk-reduction approach based on hail insurance. The studies conclude that since there is no scientific evidence on the effectiveness of the hail prevention system, there is a need to consider hail insurance options. Expenditures for irrigation are allocated to the Water Agency (subordinated to the Ministry of Environment) that is in charge of managing the large-scale central irrigation systems (CIS). The Water Agency manages 78 CIS located throughout the country, which have the (theoretical) capacity to irrigate as much as 144,000 hectares of land. In reality, only 7,000-15,000 ha are irrigated on annual basis by the use of CIS, and out of 121 operating irrigation stations only 40-50 are being utilized. So the bulk of annual budget allocations for irrigation (about 75% of total) are in fact staff costs – WA employs over 1000 people (mainly CIS operators and guards). The rest – up to 5 million MDL on annual basis – is used for maintenance and rehabilitation of irrigation systems. It is unlikely that this limited amount is sufficient to perform adequate maintenance of the CIS, much less serious rehabilitation. However, without a significant rehabilitation of the central irrigation systems, and the 45 Source: Rural Productivity in Moldova – Managing Natural Vulnerability, The World Bank, May 2007. 62 consequent expansion of irrigated areas, fees collections from water users will also be insufficient for maintenance. Earlier studies on this topic46 concluded that prospects for the Water Agency to properly operate and maintain the CIS over the longer term were limited, and recommended irrigation management transfer to WUAs as the only viable option for long-term sustainability of the CIS in Moldova. The Ministry of Environment has recently reported that a concept is presently under preparation aiming for a comprehensive reform of the irrigation management system in view of rationalizing the public spending and ensuring a better service to its clients, including CIS management transfer to WUAs, where feasible. The Advisory Service carried out by the Rural Development Agency has been created in Moldova in early 2000 with the support of the World Bank and SIDA, while the full financing of this service has been picked up by the public budget in mid-2013, when the donor funding ceased. The extension network is managed from the head office located in Chisinau and consists of 35 regional offices that involve 75 regional consultants and 350 local consultants operating within village mayoralties. The services of the network are presently provided to the farmers free of charge. The coverage and the efficiency of the extension network are estimated as high and continuously improving. Presently, the network is present in every rayon, covers over 40 percent of villages/mayoralties and over 50 percent of farms. Services are provided to all types of farms, including small subsistence farms which form the largest client group. Most of the advice offered relates to production technologies (52 percent of services in 2014), while other consultancy areas include marketing (16% of services), business (20%) and legal advice (12%). Testing and registration of new crop varieties is done by a public institution – State Commission for Testing of Plant Varieties (shortly referred to as Seeds Commission) subordinated to MAFI. The operation of this institution is mandated by Moldova’s legal requirement to test any new variety before allowing it for sale on the domestic market. So the clients of the Seeds Commission are either local breeders (mainly ag research institutes) or seed importers that want to introduce on the market a variety that is not yet registered in the national catalogue of plant varieties (for a charge). The Seeds Commission has a central office as well as regional stations/labs with adjacent land plots for carrying out the field tests (the total land area managed by the SC – 840 ha). Most of the varieties tested by the SC are imported. There is a long standing discussion in Moldova about the need to change the role of the Seeds Commission from a prescriptive to a consultative one by abolishing the mandatory requirement to test the new varieties that are used in other countries although not yet registered in Moldova. This change would better serve the needs of the private sector by offering immediate access to new modern plant varieties developed elsewhere. Under the renewed mandate the Seeds Commission would carry on the variety tests with the purpose of identifying and recommending varieties that perform best in the pedo-climatic conditions of Moldova, concomitantly not precluding the use of imported varieties. 46 Institutional Assessment of Moldova’s Central Irrigation Systems, Report prepared for Millennium Challenge Corporation, November 2008. 63 Annex 6: Taxation of Agriculture Property in Moldova Land tax Real property tax For agricultural land: For lands for agricultural For non-value based For agricultural buildings, enterprises, other than land outside of apartments, individual non-value based : residence zone houses in rural areas and non-value based property a) all land, other than a) in cities and in rural a) on which are located The tax rates are the meadows and pastures: areas - MDL 10 (€0.49) per buildings and following: 2 100 m construction, quarries - with fertility index - MDL and unevaluated lands -0.1% from “normative” 1.5 (€0.073) per grade- destroyed by production value – for natural persons; hectare activity - MDL 350 - 0.1% from book value – for (€17.08) per hectare. - without fertility index - MDL legal persons 110 (€5.37) per hectare b) meadows and pastures: b) in Chisinau and Balti - b) MDL 70 (€3.42) per MDL 30 (EUR 1,46) per 100 hectare. - with fertility index - MDL m2; 0.75 (€0.037) per grade- hectare - without fertility index - MDL 55 (€2.68) per hectare c) lands occupied by water c) in other cities and towns bodies (ponds, lakes, etc.) - residence - 10 MDL (€0.49) 2 MDL 115 (€5.61) per hectare per 100 m . aquatic 64 Annex 7: Preferential Tax Treatment in Agriculture Defining preferential tax treatment of agriculture requires an exact definition of the agriculture sector and agreement on what constitutes tax relief (concession).To define agriculture it is necessary to define who qualifies as a farmer and distinguish agricultural activity from the provision of other land-based goods and services, such as housing for farmers and others in rural areas. In broad terms, tax expenditures are concessions that fall outside a tax norm or benchmark (World Bank 2004). A useful and more detailed description is “any law … that exempts, in whole or in part, certain persons, income, goods, services, or property from the impact of established taxes, including, but not limited to, tax deductions, tax deferrals, preferential tax rates, and tax credits” (State of Oregon 2003). In this study tax expenditures are defined as tailored reductions in tax liability. Special tax incentives or tax expenditures that generate reductions in tax revenue constitute a fiscal advantage conferred on farmers individually or the agriculture sector as a whole .Tax expenditures may take the following forms:  Exemptions: amounts excluded from the tax base  Allowances: amounts deducted from the “normal” or benchmark regime to arrive at the tax base  Credits: amounts deducted from tax liability  Reduced rate: a special tax rate applied to a specific class of taxpayers or taxable transactions  Tax deferral: a relief that takes the form of a delay in paying taxes The farmer is defined in this study as an individual enterprise or operation based on private ownership of farmland and related property, with family members working personally to earn income from the production of agricultural products. Their activity must consist in the primary processing of agricultural produce, which extends to commercializing their production. The area of land being cultivated or otherwise farmed and income from renting agricultural implements and land from the individual farm should provide more than 50 percent of the farmer's annual earnings. Agricultural producers may be joint stock companies, limited liability companies, or agricultural cooperatives. Measuring the Costs of Tax Expenditures Measuring tax expenditure costs and effects is very complex. While there is broad agreement in theory on the definition of tax expenditures, there seems to be no well-accepted methodology for estimating them (Canada, Department of Finance 2001. Internationally there is a range of approaches to costing tax expenditures, with some of them quite broad in terms of what forms of concessions are covered. In principle, costing out a tax expenditure involves comparison with a benchmark tax system (what is normal), based on the methodology chosen. Estimates of tax expenditures depend heavily on the choice of the “normal” or benchmark regime against which treatment of agriculture can be compared. Thus, the choice of benchmark is more than technical; it is rooted in different views of the normative tax base (OECD 2004). There are two broad approaches to this choice: 65  The conceptual approach, in which the benchmark is defined as an abstract tax system that is seen to have desirable neutrality properties in that it does not give particular tax advantages to a limited number of people or activities.  An approach based on current tax law, but without provisions seen as benefitting particular groups of people or businesses or favoring particular activities, such as research or investment in pensions. There are also different methods for estimating the costs of tax expenditures with the following three principal methods distinguished in the literature:  The Foregone Revenue Method is an ex post quantification of the reduction in tax revenue relative to a benchmark due to tax expenditures. It does not take into account changes in the behavior of individuals.  The Outlay Equivalent Method estimates the subsidy or transfer that would leave taxpayers with an income net of taxes similar to what they would obtain if the tax expenditure applied.  The Revenue Gain Method measures how much revenue could increase if a particular tax concession were removed. Accurate estimation of this cost requires estimating the secondary or behavioral effects associated with the change. The method applied in this study was the foregone revenue approach with the estimations referring to the last year for which data are available (usually 2014) and are calculated using and current tax law benchmarks. The method is based on quantification of the decrease in revenues that would be caused if the taxpayer did not take advantage of the tax relief. In other words, this is a static calculation of lost revenues incurred by the public sector. Naturally, some simplifications are used in calculating tax expenditures in this study as quantifying the real amount of tax relief is difficult. Measurement of fiscal costs is only the first step in the process of evaluating tax expenditures. A full-fledged economic evaluation requires an even more comprehensive approach involving (1) identification of the evolution and role of a tax expenditure in relation to policy objectives and in comparison with alternative instruments; (2) measurement of the behavioural responses that are crucial for judging the effects of the relief; (3) simulation and accounting methods to generate estimates of the effects; and (4) an analysis of the formal legal rules through which relief is delivered to identify administrative and compliance costs. This study uses a simple method to measure the costs and gauge the economic effects of tax expenditures (for assessment of how tax expenditures relate to policy objectives, see the next section). 66 Annex 8: Calculating VAT GAP Quantitative assessment of the VAT gap is based on the supply and use matrices and theoretical assessment of taxes that should be collected. Several numbers are calculated on the basis of these data:  Output VAT – total output of sector s*VAT rate for sector s ( )  Input VAT credit –,the VAT that was paid by the sector and is deducted from the VAT bill (∑ input from sector s to sector r*VAT rate for sector s), ( )  Investment VAT credit (investment in sector s*VAT rate for sector s), ( )  Import VAT (total imports of sector s* VAT rate of sector s) ( )  Export VAT credit (as export is exempted from VAT in most countries to avoid double taxation) (Export of sector s * VAT rate of sector s), ( ). The total VAT that should be collected from sector s is then calculated as follows: ( ) ∑ where is the percentage of output by non-registrants. In calculating the compliance gap¸ the VAT rates that are taken into account are those implied by current legislation whereas in calculating the policy gap the basic tax rate is used. Some caveats must be kept in mind in analysing the results of such a simple analysis. First and foremost, it does not take into account the consumption shift resulting from the change of relative prices. Such effects may be quite large, since the reduced VAT rate in place for agricultural products may be replaced by home production. The second important problem is that the percentage of output that is exempt is a rather rough estimate based on tax authority data. Moreover, the percentage of output by non- registrants ( ) is also a very approximate number may be dependent on the VAT rates. Therefore, the results of such exercise should be treated as a first estimate of the tax gap rather than complex and robust estimations. 67 Annex 9: Concessions in Taxes on Incomes (unincorporated businesses) Special system applied to agriculture Other concessions (rates, thresholds, etc.) Australia Income averaging for primary producers Farm Management Deposit Scheme (stabilizing measure) Deferral of income from double wool clip Spreading income from insurance recoveries Deferral or spreading of income from forced disposal or death of livestock Valuation of livestock from natural increase Income tax exemption for dairy exits Exemption of Sugar Industry Exit Grants Deductions for various activities (e.g., accelerated depreciation for water management costs) Austria Liability is based on assessed value of the Cash accounting for smaller farms farm, which is lower than market value. Below € 150, 000 assessed value, tax is based on standard calculations Belgium Flat rate system used to determine what Specific tax allowance applied to subsidies to constitutes taxable income; widely used in agriculture from the European Commission agriculture but also for other professions Germany Method of calculation: four systems Agricultural income allowance determined by turnover and organizational Income tax reduction under the simplified scheme form, based on (a) accounts – 30%; (b) Simplified calculation of gross agricultural income simplified accounts - 15%; (c) standards – 50%; (d) estimates by authorities – 5%. France Type of valuation used to determine farm Income smoothing mechanisms income based on farm type and department Reduction of 50% applied to young farmers in the (>50% of farmers). first 5 years of farm takeover. Ireland Tax averaging Accelerated capital allowances (pollution control measures) Concessions to young trained farmers on stock relief. Concessions to retiring farmers Italy Type of valuation used to determine farm Japan income based Agricultural on farm income taxtype and region Rate lower than on wages and salaries (about 50%) Korea Special agricultural income tax, but structured Nether- like the general system. Income averaging not specific to agriculture lands Forestry income exempt New Income equalization scheme, also covers fishing Zealand and forestry Norway Special allowance against taxable income up No tax averaging in agriculture, but averaging for to NOK 61,500 forestry and reindeer 68 Poland Only production of a specific list of products is covered by the tax system. Taxable income may be assessed using a standard rate or on Slovak the basis of accounts Previous reduced rates of income tax for farmers, Republic tax holidays for new entrants and special treatment of purchasers of state assets have now been phased out. United Tax averaging Kingdom United “Family-farm corporations” can use the cash Sole farmers and partnerships may form three-year States method if their taxable gross incomes are less average Joint income; on Committee this is only available Taxation also liststo farmers cash than US$25mn (c.f., US$1 mn for others). and on farm accounting income. for specialenvironmental National farmers, treatment of certain and are subsidies costs sometimes carry-back tax-free for the farmers. of expenses. 69 Annex 10: Fundamental Principles of a Fair and Effective Tax System An effective tax system is equitable, supports economic growth, and has minimal impact on taxpayer behavior. It should be based on several general principles and take into account the specific objectives and characteristics of each sector. The three fundamental tax principles for a fair and effective tax system are  Equity: The tax system must treat taxpayers in different sectors of the economy equitably. People with similar incomes should pay similar rates—horizontal equity. There is also widespread agreement that vertical equity should apply: people with higher incomes should on average pay a higher effective tax rate.  Incentive: The tax system should facilitate and support economic growth and job generation. In particular, it should not operate as a disincentive for those with low incomes to become employed or as a disincentive to enterprise, innovation, and investment. Finally, it should not distort production and trade.  Economic Neutrality: This term covers a number of characteristics that ensure that the behavior of taxpayers is not unduly influenced: (1) The tax base should be as broad as possible and tax rates as low as possible. (2) Taxation of income from different sources—employment, self- employment, investment, pensions—should be taxed equitably. (3) Administrative and compliance costs should not be excessive. 70 Annex 11: Reduced VAT rates for food and agricultural products – pros and cons There is a number of economic arguments for granting lower rates to very specific sectors, e.g. agriculture. The pros and cons of these arguments can be summarized as follows47: 1. Compliance cost. Extending reduced VAT rates to food, as Moldova does, affects the efficiency of a VAT system and increases complexity, which in turn increases the administrative costs (for system operation and monitoring) and compliance costs. The latter seem to be particularly high for the food sector due to the multitude of products and the gray zone between sales of food and of prepared food served in restaurants. Reduced rates on food also tend to make do-it- yourself cooking economically more attractive than visiting restaurants. 2. Higher productivity and employment. Permanently lowering the VAT rate on a particular good or service is expected to lead to its lower price, which in turn translates, as evidence suggest, into increased consumer demand for this particular good or service. Thus, production and employment in the sector producing that good will correspondingly expand to meet the higher demand. However, the strength and swiftness of the production and employment response depend on consumer response to lower prices and the degree of competition within the sector. If consumption is price-inelastic—typically the case for basic goods, such as food— production and employment will not increase much. An empirical study of the EU countries found that food is price-inelastic: a drop of 1 percent in food prices expands food consumption by less than 0.5 percent. In sectors where there is little competition, pass-through to prices may be less than full, which mutes the impact on production and employment. Second, there is a convincing argument for extending reduced VAT rates to sectors whose services are easily substituted for do-it-yourself or underground work (EU 2009), and that is that high VAT rates make it very expensive to buy these services or products on the market and more attractive to supply oneself. The implication is that high-skilled professionals spend time on low- skilled work at home rather than spending time with their families or increasing their more productive labor supply.48 Third, there is a theoretical but not an empirical argument for extending reduced VAT rates to sectors that employ many low-skilled workers in order to boost demand for them. In theory, by boosting demand for agricultural products, reduced VAT rates stimulate demand for low-skilled workers and push up their wages so that employment becomes a more attractive option than unemployment. In practice, the impact will be more pronounced the larger the share of low-skilled jobs in the favored sector relative to the rest of the economy, especially if it is a non-tradable sector that is not very vulnerable to competition from abroad. If like agriculture the sector is tradable, part of the increase in demand would go to increasing imports, boosting low-skilled employment abroad 47 See Copenhagen Economics (2007). 48 This argument, however, does not apply to the food products that were granted reduced rates in Moldova. It is difficult to imagine that high market prices will induce skilled professionals to start producing their own milk or sugar. 71 rather than at home. Again, empirical work on EU countries found evidence that the overall impact on demand for low-skilled workers stemming from reduced VAT rates for food product is not impressive. 3. Better distribution of income. Lower VAT rates on essential goods are typically introduced to ensure that the items are more affordable for lower-income households. Reducing VAT rates on food that constitutes a larger share of consumption for low-income than for high-income households implies a cost saving. The larger the difference in consumption shares is, the more effective the argument. Unfortunately, the evidence shows that the argument has empirical support only in countries where income inequality is high, which is not true of Moldova. As such measures are universal, however, they also imply that the discount is also enjoyed by better-off households. Furthermore, to the extent that better-off households consume more in absolute terms than poorer households, higher-income households might be receiving a higher share of the VAT subsidy than lower-income ones. Except for Denmark, all advanced EU countries have lower VAT rates for food and agricultural products, with the tax rates ranging from 0 percent in Ireland and the U.K. to 15 percent in Norway. However, in the new member states, the use of reduced VAT rates for food products is less common. Among Moldova’s benchmark countries, using reduced VAT rates and standard rates are equally popular, but in general, standard rates in these countries are lower than those in the EU, so the burden on producers is less onerous. In addition, simplified or special VAT regimes for agriculture are used by many countries in the EU (usually available to small farms), as well as in Ukraine49. Such special VAT regimes allow agricultural producers to accumulate VAT at special accounts and spend these funds for production purposes without transferring them to the budget. 49 At present the special VAT treatment of agriculture products in Ukraine is not referring to small farms. However, the government of Ukraine is now in the process of bringing preferential VAT treatment of agriculture closer to the general VAT regime (as a step to harmonize it with the EU standards). 72 Country VAT Rate Country VAT Rate “Old EU” and EFTA New EU Member States Austria 10% (reduced) Bulgaria 20% (standard) Belgium 6% (reduced) Croatia 5%/13% (reduced) Denmark 25% (standard) Czech Republic 15% (reduced) Finland 14% (reduced) Estonia 20% (standard) France 5.5% (reduced) Hungary 18% (reduced) Germany 7% (reduced) Latvia 21% (standard) Greece 13% (reduced) Lithuania 21% (standard) Ireland 0% (most foodstuffs, except confectionery), Poland 5%/8% (reduced) livestock taxed at 4.8% Italy 4% (reduced) Romania 24% (standard) Luxembourg 3% (reduced) Slovakia 20% (standard) Netherlands 6% (reduced) Slovenia 9.5% (reduced) Norway 15% (reduced) Portugal 6% (reduced) Spain 4% (basic foodstuffs), 10% (other) Sweden 12% (reduced) Switzerland 2.5% (reduced) United Kingdom 0% (reduced, certain foodstuffs) Benchmark countries Armenia 20% (standard) Moldova 8% (reduced) Azerbaijan 18% (standard) Montenegro 7% (reduced) Belarus 10% (reduced) Serbia 10% (reduced) Georgia 18% (standard), with many exemptions Ukraine 17% (standard), with special tax regime for agriculture Macedonia, FYR 5% (reduced) Sources: 1. VAT Rates Applied in the Member States of the European Union, 2015, European Commission, http://ec.europa.eu/ taxation_customs/resources/documents/taxation/vat/how_vat_works/rates/vat_rates_en.pdf; 2.Worldwide VAT, GST and Sales Tax Guide 2014, EY, http://www.ey.com/Publication/vwLUAssets/Worldwide-VAT-GST-and-sales-tax-guide-2014/$FILE/Worldwide-VAT-GST-and-sales- tax-guide-2014.pdf; 3. http://www.vatlive.com/vat-rates/international-vat-and-gst-rates/. 73 Annex 12: Distributional Impact of a VAT Increase for Agricultural Products Lower-income households are likely to suffer due to the squeeze on their budgets that would result if the VAT rate on agricultural products went up from 8 to 20 percent. To capture these effects, as table B1 shows, simulated different combinations of changes of the VAT have been simulated and their poverty impacts analyzed. The analysis is based on simulating in the micro-data (HBS 2013) the purchasing power loss experienced by each household, based on its consumption patterns before the VAT increase. The analysis does not consider possible substitutions between goods but shows two different assumptions about the pass-through of the price increase. On the assumption that the entire increase in the VAT would be passed on to consumers, the poverty impacts or raising the VAT on all categories of agricultural products is 0.6 percentage points, which is roughly equivalent to 22,000 more people becoming poor. A lower pass-through would restrain this poverty impact. Interestingly, increasing the VAT on agricultural products, dairies, and grains would have the same poverty impact as abolishing all the other exceptions on the VAT. Table B1, Moldova: Poverty Rates under Different Assumptions about the VAT rate and the Pass-through of Increases Pass-through=1 Pass-through=0.8 Baseline Poverty rate in 2013 12.7 12.7 Simulations (1) VAT increase for agricultural goods 13.0 13.0 (2) VAT increase for food (selected categories) 13.0 12.9 (1)+(2) VAT increase for both agricultural goods and 13.3 13.2 Source: HBS 2013. food (selected categories) Urban and rural areas will be affected differently. Urban areas, which in the baseline start with a much lower poverty rate (4.6 percent, compared to 18.8 percent in rural areas) would see only a marginal increase in poverty of an 0.2 percentage point from the combined effect of raising the VAT on agricultural products, dairy, and grains. In contrast, in rural areas the increase in poverty incidence would amount to 1 percentage point. Given that the poverty impacts of raising the VAT rate are not negligible, there is scope for reallocating resources saved by abolishing the reduced VAT rate to compensate poor and vulnerable groups. TableB.2 shows in absolute terms and as a share of the estimated current expenditure on the reduced VAT rate, the costs of compensating different groups. It presents results for three groups, though many others would be possible. The groups that could be given priority for compensation are: (1) beneficiaries of Adjudor Popular, the current Guaranteed Minimum Income program, which could add a top-up to compensate for the increase in the VAT; (2) the poor, on the ground that protecting them from further economic hardship should be a priority; and (iii) the poor and those who are vulnerable to poverty due to the VAT increase. 74 Table B.2: Alternative Compensation Scenarios, VAT Increase for Agricultural Products Percentage of the Cost of the Groups compensated in each scenario Total Subsidy Adjudor popular beneficiaries 1,483,839 2 Poor households 3,899,164 5 Poor households and households that become poor before the VAT increase 4,064,945 5.4 but became poor after Source: World Bank staff These are rough scenarios in that they assume that the targeted groups can be identified perfectly and that there are no administrative costs in delivering the targeted subsidy, but they do give a sense of the fiscal savings that could be obtained from replacing a universal subsidy with a targeted one. In particular, adding a top up to Adjudor Popular, even as Moldova is seeking to expand its coverage, would absorb the least resources, both because the program reaches only a limited number of people and because the administrative costs of a top-up would be minimal, since the whole targeting and administrative system is already in place. Reaching the other two groups would be more complicated, so that the estimated costs as a share of resources underestimate the true resources required. Still, because of the targeted nature of the compensation suggested, less than 6 percent of the budget currently absorbed by the VAT discount would be absorbed by compensation of all the poor and vulnerable. 75 Annex 13: Tax Incentives and Investment in SMEs Most modern investment models are usually based on the neoclassical theory that a representative firm maximizes the present value of its expected profits. The two most common models following this tradition are the theory of the user cost of capital and the Q-theory. The basic reasoning behind the former, introduced by Jorgenson (1963) and Hall and Jorgenson (1967), is that a firm decides to invest when marginal benefits exceed marginal costs. For this model the cost of investment is a function of the returns required on debt and equity and an adjustment for corporate taxes. Thus, corporate taxes on profits increase the cost, while allowances for depreciation and investment tax credits reduce it. This relation can be expressed as follows: (Y/K)(1 – u) = (r + d)(1 – A) [1a] or equivalently, Fk = (r + d)(1 – A)/(1 – u) 1b In equation 1b, the term Fk = (Y/K) represents the increase in gross revenues (Y) accompanying a (one currency) unit increase in the representative firm’s capital stock (K). With diminishing returns to installed capital at the margin, Fk falls as the capital stock increases. Revenues from investment at the margin are subject to the statutory or “headline” CIT rate, denoted by (u). The left -hand-side of 1a measures the after-tax marginal benefit from an additional unit of investment. This framework is useful for considering the channels through which various tax incentives may operate to encourage investment. First, reducing the statutory CIT rate will increase the after-tax revenues from investment at the margin, which tends to lead to a higher equilibrium for capital stock. A reduction in the CIT rate, however, also lowers the present value of deductible depreciation allowances, which lowers A. A reduction in the CIT rate also increases the after-tax cost of debt financing by reducing the value of interest deductions, which also acts to lower A. Therefore, the impact on investment incentives of a reduction in the CIT rate is ambiguous. However, the first-noted effect will generally dominate with typical parameters, implying that investment incentives will be increased by a reduction in the corporate tax rate. Second, introducing or enriching a system of investment tax credits increases the value of A, which tends to encourage investment at the margin. Similarly, increasing the rate at which capital can be depreciated for tax purposes (e.g., accelerated depreciation, or immediate and full expensing of capital costs) increases A and thereby investment incentives. Depending on the rate and design of the investment tax credit and capital cost allowance (i.e., tax depreciation) regime, the term (1 – A) may be negative and the tax system on balance may operate to encourage rather than discourage investment relative to the no-tax case. Such situations lead to an increased benefit when costs rise and so are particularly susceptible to tax avoidance and inefficient investment. Third, government policy can influence the firm’s pre-corporate tax cost of finance (r). As already noted, the cost of finance, which generally is some weighted average of equity and debt finance, will tend to increase if the statutory CIT rate is reduced. In some cases, the cost of equity financing may be a 76 function of personal tax parameters, such as the degree of relief from double taxation (corporate and personal tax integration). In particular, reductions in shareholder dividend tax rates and capital gains tax rates may lower the cost of funds for SMEs that obtain financing from domestic rather than international capital markets. Source: OECD (2009). 77 References Bravo-Ortega, C. Lederman, D. 2005. "Agriculture and national welfare around the world: causality and international heterogeneity since 1960,"Policy Research Working Paper Series 3499, The World Bank. Budget Officials. GOV/PGC/SBO(2004)6 Cimpoies, D., Lerman, Z., Racul, A. 2009. The economics of land consolidation in family farms in Moldova. European Association of Agricultural Economists, 111th Seminar, June 26-27, 2009, Canterbury, United Kingdom, http://ageconsearch. umn.edu/handle/52837 Cimpoies, D., Muravschi, A., Racul, A. 2008. Structural changes in Moldovan agriculture: problems and perspectives. Poster Paper, IAMO, http://www.iamo.de/ uploads/media/2b.3_Cimpoies_CD.pdf COPENHAGEN ECONOMICS. (2007). Study on Reduced VAT Applied to Goods and Services in the Member States of the European Union. Retrieved from . Csaki, C. (2008): A joint IAAE-EAAE seminar summary comments. In: Csáki, C. and C. Forgács (eds.): Agricultural economics and transition: What was expected, what we observed, the lessons learned. Proceedings Vol. 1. Studies on the Agricultural and Food Sector in Central and Eastern Europe, Vol. 44: 3- 19. Leibniz Institute of Agricultural Development in Central and Eastern Europe (IAMO), Halle (Saale). Csaki, C., Kray H. (2005). The Agrarian Economies of Central-Eastern Europe and the CIS: An Update on Status and Progress in 2004. ECSSD Working Paper No. 40, Washington, DC: World Bank (2005). IMF (2011). Revenue Mobilization in Developing Countries. IMF Fiscal Affairs Department Imf (2014) Hutton, E., Thackray, M., Wingender P. Revenue Administration Gap Analysis Program—The Value-Added Tax Gap. IMF (2009A) Klemm, A., Van Parys, S. Empirical Evidence on the Effects of Tax Incentives. IMF. WP/09/136 Lerman, Z., and Sutton W. (2008). “Productivity and Efficiency of Small and Large Farms in Moldova.” Post-Soviet Affairs 24(2): 1-24. Martin, W., Mitra, D. (2001). Productivity Growth and Convergence in Agriculture versus Manufacturing. Economic Development and Cultural Change, 49(2), 403-422. Mundlak, Y. (2000). Agriculture and Economic Growth: Theory and Measurement. Cambridge Massachusetts: Harvard University Press. OECD (2004) Best Practice Guidelines. Off-Budget and Tax Expenditures. Working Party of Senior OECD (2005) Taxation and Social Security in Agriculture. ISBN 92-64-01364-4. www.oecd.org 78 OECD (2006): The new rural paradigm: Policies and governance. Organization for Economic Co-operation and Development (OECD), Paris. OECD (2007). Tax Incentives for Investment – A Global Perspective: experiences in MENA and non-MENA countries OECD (2009), Taxation of SMEs: Key Issues and Policy Considerations, OECD Tax Policy Studies, No. 18, OECD Publishing, Paris. OECD (2014), Policy Framework for Investment in Agriculture, OECD Publishing. http://dx.doi.org/10.1787/9789264212725-en Restuccia, D., Adamopoulos, T. 2012. "Land Reform and Productivity: A Quantitative Analysis with Micro Data,"2012 Meeting Papers 1083, Society for Economic Dynamics. Rozelle, S., Swinnen, J. (2009). "Political Economy of Agricultural Distortions in Transition Countries of Asia and Europe" Agricultural Distortions Working Paper 50298, World Bank. Swinnen, J.F.M., Vranken L (2010), Review of the transitional restrictions maintained by Bulgaria and Romania with regard to the acquisition of agricultural real estate, CEPS Special Report prepared for the European Commission, CEPS, Brussels. Swinnen, J.F.M., Vranken L. (2009), Land & EU accession: Review of the transitional restrictions by new member states on the acquisition of agricultural real estate, CEPS, Brussels. Tax by Design: the Mirrlees Review, J. Mirrlees, S. Adam, T. Besley, R. Blundell, S. Bond, R. Chote, M. Gammie, P. Johnson, G. Myles and J. Poterba, ISBN: 978-0-19-955374-7, Oxford University Press: September 2011. World Bank (2003) Why worry about tax expenditures? PREM notes number 77. Washington, The World Bank. World Bank (2007a) Moldova: Agricultural Research, Education and Training Study, Washington, DC: The World Bank. World Bank (2007) Rural Productivity in Moldova – Managing Natural Vulnerability. Washington, DC: The World Bank. World Bank (2004) Tax Expenditures. Shedding Light on Government Spending through the Tax System: Lessons from developed and transition economies. Ed. Brixi, H. P., Valenduc, C. M. A. and Swift, Z. L. Washington, The World Bank. ISBN 0-8213-5601-1 World Bank (2008). World Development Report 2008: Agriculture for Development. Washington, DC: The World Bank. 79 World Bank (2013) Reducing the Vulnerability of Moldova's Agricultural Systems to Climate Change:Impact Assessment and Adaptation Options William R. Sutton, Jitendra P. Srivastava, James E. Neumann, Ana Iglesias and Brent B. Boehlert World Bank Studies. October 2013 World Bank (2013). James, S., Tax and non-tax incentives and investment: evidence and policy implications. Investment Climate Advisory Services. 80